* Update dotnet-mcp-builder skill to ModelContextProtocol 2.x
Align the skill with the C# SDK 2.0.0 release and the MCP 2026-07-28
spec: stable line is now 2.x, HttpServerTransportOptions.Stateless
defaults to true, roots/sampling/MCP-channel logging are [Obsolete]
(MCP9005) with the multi-round-trip input_required pattern as the
replacement, discovery-first negotiation (server/discover) supersedes
the initialize handshake, Mcp-Method/Mcp-Name routable headers, raw
structuredContent for non-object results, required Tool.inputSchema,
and the new ModelContextProtocol.Extensions.Tasks and
ModelContextProtocol.Extensions.Apps packages (typed MCP Apps support
replacing the hand-rolled _meta/ui:// pattern on 1.x).
* Address Copilot review: Apps extension accuracy, header scope, capability ownership
- packages.md: the Apps package replaces the manual _meta wiring, not the
ui:// resource; note the experimental MCPEXP003 diagnostic; label the
1.x -> 2.0 list as highlights and add the OAuth/SSE runtime changes
with a pointer to the full release notes.
- transport-http.md: Mcp-Method is on every POST, Mcp-Name only on named
invocations (tools/call, prompts/get, resources/read) - do not require
it globally at gateways.
- mcp-apps.md: current MIME type is text/html;profile=mcp-app (skybridge
is a legacy draft value); document [McpAppUi] + WithMcpApps().
- server-features.md: roots/sampling are client capabilities, only
logging sits on ServerCapabilities.
* Correct stateful HTTP guidance: 2026-07-28 has no HTTP sessions
Per the official SDK v2 elicitation docs, a server with Stateless=false
refuses the 2026-07-28 revision so dual-path clients fall back to an
initialize-capable revision; ElicitAsync cannot be used on 2026-07-28
Streamable HTTP at all. Reframe stateful HTTP as down-level
compatibility mode and document the multi-round-trip pattern
(InputRequiredException / InputRequest.ForElicitation, retry with
InputResponses -> ElicitResult) as the current-protocol way to ask
mid-tool, across SKILL.md, transport-http.md, and elicitation.md.
* Update typescript-mcp-server-generator skill to MCP TypeScript SDK v2
Replace the retired monolithic @modelcontextprotocol/sdk with the v2
focused packages (server, node, core, framework adapters), require
zod@^4.2 and Node 20+, document the registerTool config-object API,
the ctx handler context, the new error hierarchy, removed SSE/WebSocket
transports, and the v1-to-v2 codemod migration path.
* Address Copilot review: adapter peer frameworks and sampling consistency
Framework adapters now note their required peer framework install
(e.g. @modelcontextprotocol/express + express), and the two remaining
sampling recommendations are replaced with the multi-round
input_required pattern that v2 recommends over the deprecated
sampling subsystem.
* feat: add Trojan Skill Hunter agent for auditing AI customization contributions
Adds a new agent that audits .agent.md, SKILL.md, .instructions.md,
hooks.json, and MCP/plugin manifests for hidden prompt injection, MCP-style
tool poisoning/shadowing, unicode steganography, excessive-agency scope
mismatches, and rug-pull/supply-chain drift before contributions are
merged, installed, or trusted.
Mapped to the OWASP Top 10 for LLM Applications (2025) and Invariant Labs'
MCP Tool Poisoning Attack research. Includes an explicit self-defense rule
so the agent treats reviewed content as untrusted data, never instructions
to obey.
* fix: reword attack-pattern examples to avoid PR risk-scan false positives
The Rule Zero bullet and two rug-pull/unpinned-fetch examples quoted the
exact literal phrases (e.g. 'ignore previous instructions', 'curl | bash')
that the repo's automated PR Risk Scan greps for. Reworded to convey the
identical meaning without the literal trigger strings - confirmed locally
with 'node eng/pr-risk-scan.mjs' (0 high/medium/info findings, was 3 high).
---------
Co-authored-by: Shubham Jiyani <shubham.jiyani@atqor.com>
* Add azure-well-architected-review skill
Adds a skill that performs a structured Azure Well-Architected Framework
review of a workload's IaC (Bicep, Terraform, ARM) and deployed
infrastructure across the 5 WAF pillars, then creates GitHub issues to
track remediation. Mirrors the existing aws-well-architected-review
skill, which had no Azure equivalent.
* Address Copilot review feedback on azure-well-architected-review
- Step 2: always inventory live Azure resources and compare with IaC to
surface drift, instead of scanning IaC only
- Step 5: add an explicit confirmation gate before any GitHub issue
creation; ambiguous or negative responses stop with a console report
- Fix nested code fences in the issue-body template by using a
four-backtick wrapper
* Add gitmoji skill and gitmoji-setup agent
Adds two complementary artifacts for the gitmoji commit convention
(https://gitmoji.dev):
- skills/gitmoji: generates gitmoji commit messages from a diff, staged
changes, or a change description. Message-only by design (never runs
git commands), with disambiguation rules and a full reference table of
the 75 official gitmojis generated from the official gitmojis.json.
- agents/gitmoji-setup: sets up gitmoji tooling in a repository. Audits
the existing hook manager and commit convention, then installs either
a non-interactive prepare-commit-msg prefill hook (default, works in
GUI clients and CI), the gitmoji-cli interactive picker, or commitlint
enforcement, without clobbering existing hooks.
Generated README indexes updated via npm start.
* Use local commitlint binary instead of npx in verify step
Addresses the package-exec-command finding from the PR risk scan: the
verification example now calls the locally installed
./node_modules/.bin/commitlint rather than npx, which could fetch and
execute a package on the fly.
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* Address Copilot review feedback
- Quote the agent description in single quotes per AGENTS.md convention
- Resolve the effective hooks directory via git rev-parse --git-path
hooks for both audit and installation, instead of hard-coding
.git/hooks (core.hooksPath, linked worktrees)
- Correct the prefill-hook compatibility claim: it prefills only when
the message editor opens and silently no-ops for -m/-F, GUI message
boxes, and CI
- Match the official gitmoji set explicitly when detecting an existing
emoji, instead of treating any non-ASCII start as one
- Drop .txt from the docs heuristic (it shadowed requirements.txt) and
remove the dependency-manifest fallback entirely: filenames cannot
distinguish upgrade/add/remove/pin/downgrade
- Restrict gitmoji -i to repos whose effective hooks dir is .git/hooks;
wire the picker through the hook manager otherwise
- Merge gitmoji into an existing commitlint config instead of
overwriting commitlint.config.mjs
- Fix the verification sequence: clean starting state, non-colliding
scratch file, abort by clearing the editor, explicit unstage/remove/
switch-back/branch-delete cleanup
- Skill: ask the user for commit history instead of running git log,
honoring the message-only contract
* Address second round of Copilot review feedback
- Pair the prefill hook with a commit-msg guard: prefilling an empty
COMMIT_EDITMSG defeats git abort-on-empty-message, so an untouched
prefill would create a commit named only with the emoji. The guard
rejects messages that contain nothing but the prefilled gitmoji.
- Extract the gitmoji alternation into a GITMOJI_RE variable shared by
both hooks.
- Document the commitlint-config-gitmoji format mismatch: it enforces
the hybrid <gitmoji> type(scope?): subject format and rejects the
plain gitmoji format produced by Options A/B and the gitmoji skill.
Option C now asks the team to choose a format first, and the
verification example uses a valid hybrid message.
---------
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
The instruction tells Copilot to verify PyPI/npm package and symbol names
against the attester.dev existence oracle (free keyless tier, 25 calls/day
per client IP) before installing or importing, catching hallucinated
dependency names (5.2-21.7% of LLM suggestions per USENIX Security 2025).
The preToolUse hook enforces the same check on code the agent writes:
stdlib-only script, blocks only on a confident oracle negative, fails open
on quota or network trouble, caches answers on disk. README tables
regenerated with npm start.
Syncs with jennyf19/the-workshop PR #8 (merged). Adds the
workshop-create skill with Path A (existing dir) and Path B
(new GitHub repo), explicit guard against repo-in-repo nesting.
TA agent updated with workshop-create section.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 26faf13e-639c-4a21-ac05-c0dc2bff7c62
The awesome-copilot validator (eng/validate-plugins.mjs:107-148)
strips .md from the plugin.json path and appends .agent.md to
find the source file. The rename to workshop-ta.md broke this
convention. Reverting to workshop-ta.agent.md so validation,
materialization, and README generation all work correctly.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 26faf13e-639c-4a21-ac05-c0dc2bff7c62
The .agent.md suffix leaked into the component ID registered
with the marketplace. Renaming gives a clean 'workshop-ta'
identifier while the frontmatter display name stays unchanged.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 26faf13e-639c-4a21-ac05-c0dc2bff7c62
- plugin.json: use ./agents/workshop-ta.md path format (matches all other plugins)
- workshop-ta front matter: name 'Workshop TA' preserves acronym (was 'workshop-ta')
- signal-write: add subtype field (hands-up/blocked/done/checkpoint/partnership)
so dashboard consumers can distinguish specific signal states
- npm run build: regenerated docs/README.agents.md
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 26faf13e-639c-4a21-ac05-c0dc2bff7c62
- Add YAML front matter to workshop-ta agent (name + description)
- Fix agent path: use 'workshop-ta' not './agents/workshop-ta.agent.md'
- Sort skills alphabetically in plugin.json
- Make Cairn dashboard reference conditional (full plugin from source repo)
- Update signal-write: write JSON to .signals/ AND note in journal
- Add partnership signal type to signal-write skill
- Inline CAIRN disposition in agent (treat external CAIRN.md as optional)
- Run npm run build to regenerate marketplace.json and docs
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 26faf13e-639c-4a21-ac05-c0dc2bff7c62
* feat: add convert-excel-to-md, convert-pdf-to-md, and convert-word-to-md skills
Add three new agent skills that convert common document formats to Markdown
using bundled Python scripts powered by MarkItDown:
- convert-excel-to-md: Converts .xlsx workbooks to Markdown with per-sheet
tables and embedded image extraction via a bundled Python script.
- convert-pdf-to-md: Converts .pdf documents to Markdown with text/table
extraction and embedded image extraction via PyMuPDF.
- convert-word-to-md: Converts .docx documents to Markdown with proper
image extraction replacing MarkItDown's base64 placeholders.
Each skill includes:
- SKILL.md with detailed usage instructions, output structure docs, and
a troubleshooting table
- scripts/ with the conversion Python script and requirements.txt
- references/setup.md with environment setup instructions
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: remove Markdown image syntax from convert-word-to-md SKILL.md
The CI valid-refs linter flagged the literal Markdown image syntax
containing a data URI as an invalid file reference. Replaced it with
a plain text description of the placeholder format.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: update pip install command to use scripts/requirements.txt for setup
* fix: clarify installation instructions for requirements-file in setup documentation
* fix: correct indentation in image extraction function
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* feat: add comprehensive documentation and setup instructions for convert-to-md skills
* fix: add trailing slashes to skill paths in plugin.json
* fix: reorder skills in plugin.json for consistency
* fix: update plugin.json and README.md for clarity and consistency
* feat: add convert-to-md plugin and update related documentation
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* fix: enhance conversion skills to handle mixed file types and improve error handling
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
Add a skill covering Helidon SE and MP patterns, Helidon 3 to 4 API migration guidance, and examples for HTTP, service, data, testing, security and observability in Java 21+ projects.
* Add tm7-threat-model skill for valid TM7 file generation
Adds a skill that generates valid Microsoft Threat Modeling Tool (.tm7)
files using the correct WCF DataContractSerializer format, with a minimal
reference file. Includes STRIDE threat generation workflow and a checklist
of common serialization mistakes that corrupt .tm7 files.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review: sanitize reference model, fix dangling refs, move to assets
- Move example-minimal.tm7 into assets/ per repo convention (aaronpowell).
- Remove personal/corporate metadata from the reference model (Owner,
Contributors, ChangedBy domain account, project names).
- Make the reference model self-contained: add a second stencil so the data
flow and threat SourceGuid/TargetGuid/FlowGuid all resolve to real elements.
- Fix SKILL.md threat contract (KeyValueOfstringThreatpc_P0_PhOB with
b:-prefixed KnowledgeBase fields) to match the bundled reference.
- Reconcile guidance: MetaInformation/Notes/KnowledgeBase are valid schema
elements and must be preserved; only SecurityGaps/Mitigations are invalid.
- Regenerate docs/README.skills.md and normalize line endings to LF.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address 2nd review: correct KnowledgeBase placement, valid TypeIds, z:Id uniqueness
- Document KnowledgeBase as a top-level sibling after ThreatMetaData (not embedded)
- Update skeleton to show <ThreatMetaData/>, sibling <KnowledgeBase>, and <Profile>
- Replace TypeIds absent from the bundled KB: AzureCosmosDB -> AzureSQLDB,
HumanUser -> Mobile, GenericDataFlow -> Request
- Require z:Id uniqueness across the file in the common-mistakes checklist
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* ci: exclude *.tm7 exports from codespell
MTM DataContract .tm7 exports embed base64 icon blobs whose substrings
(oT, bu, wth, mKe, ...) trigger codespell false positives. Skip *.tm7,
matching the existing convention for binary/asset files.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: prasad <prasaddharaskar@gmail.com>
Co-authored-by: Aaron Powell <me@aaron-powell.com>
- Update migration agent guidelines to prioritize extension tool usage for code migration.
- Refine migration phases with detailed steps for pre-migration review and schema migration.
- Add new reviewing skill references for PostgreSQL materialized view refresh and UNION ALL planner risks.
- Ensure consistency in collation handling and testing strategies across skills.
Co-authored-by: TCPrimedPaul <paul.delannoy@tc.gc.ca>
* Add arch plugin (architecture + modernization skills)
Add the `arch` plugin with two skills:
- `arch:document` — produce a single, cited architecture document for a
locally-cloned repo, reading files on disk only.
- `arch:modernize` — generate a phased modernization plan, auto-running the
document workflow first when no architecture doc exists.
Skill sources live in top-level skills/ and are referenced declaratively
from plugins/arch/.github/plugin/plugin.json, per the repo's plugin model.
Regenerated docs/README.plugins.md, docs/README.skills.md and marketplace.json.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Merge arch skills into single doc-and-modernize skill
Collapse the document and modernize skills into one standalone skill
(doc-and-modernize) with Documentation and Modernization modes, keeping
the plugin named arch. Modernization mode now runs the Documentation
workflow inline instead of invoking a separate arch:document skill,
fixing standalone-install cross-skill references. Reframe Documentation
mode as local-first (remote/API lookups are a flagged last resort)
rather than local-only. Regenerate docs and marketplace.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review: list skill by bare ID in arch README
Use 'doc-and-modernize' (repo convention) instead of the namespaced
'arch:doc-and-modernize', noting it surfaces as arch:doc-and-modernize
when installed via the plugin.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review: use non-HTML placeholder in instructions template
Change plain-text <PROJECT NAME> to [PROJECT NAME] in the header and
first paragraph so Markdown renderers don't parse it as an HTML tag and
drop it. The code-span `<N>` on the phase line is unaffected.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review + fix codespell CI failure
- Fix codespell: pre-empts -> preempts in the instructions template
- Consistent terminology: replace 'research step/workflow' with
'Documentation mode' in SKILL.md, README, and plugin.json description
- Fix run-on: add 'that' before 'Modernization mode must surface'
- Regenerate docs/marketplace for the updated plugin description
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review feedback
- Instruct redacting credentials/tokens from git remote URLs before recording
- CI enforcement: ask user or mark [UNVERIFIED]; remote lookup is flagged last resort
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: samqbush <samqbush@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Renames the skill so its name makes the source-code-review scope explicit and
distinct from the existing config-focused `mcp-security-audit` skill (the
duplicate-detection check flagged the shared `mcp-security-` prefix; it is
advisory only). Also sharpens the description's first line to lead with
"implementation source code of MCP servers, clients, and tool handlers".
- Rename folder skills/mcp-security-baseline -> skills/mcp-implementation-security-review
- Update name field and H1 heading to match
- Regenerate docs/README.skills.md
Co-authored-by: Swetha Kumar <16600902+Swethakumar1@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add foundry-hosted-agent-copilotkit skill (development-focused)
Reworked from PR #2090 feedback: drops all scaffolding/sample-template
content and focuses on ongoing development with CopilotKit + AG-UI +
Azure AI Foundry hosted agents — adding/gating tools, human-in-the-loop
approvals, generative UI and shared state, event-stream debugging,
pre-1.0 dependency upgrades, and the hosted-agent deploy loop.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Add HITL code examples to foundry-hosted-agent-copilotkit skill
Address review feedback: the .NET approval paragraph was prose-only.
Add a .NET snippet from the official Step04_HumanInLoop sample and
correct the type/behavior (ToolApprovalRequestContent; convert—not
remove—the request_approval call/result). Also add live-verified
Python snippets for hosted-agent approval-forwarding (#6652) and the
previous_response_id #6851 guard.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* Add mcp-security-baseline skill
An Agent Skill that reviews MCP server and client source code against a security
baseline (5 controls, 7 RCE vectors, OWASP MCP Top 10) and produces a compliance
report with file/line evidence. Complements mcp-security-audit, which checks
.mcp.json configuration.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review feedback: remove ignored keywords field, link MCP spec
- Remove the top-level `keywords` field: it is not a recognized skill
front-matter field and is ignored (skills are indexed for search by
name + description only). The keyword terms are already present in the
description, so discovery is unaffected.
- Link the MCP 2025-03-26 specification at the protocol-version check so
the agent has it for quick reference.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review feedback: fix unsafe-deserialization example, link current MCP spec
- Change the unsafe-deserialization example from `yaml.load(..., Loader=yaml.FullLoader)`
to `Loader=yaml.UnsafeLoader` to reflect genuinely unsafe usage.
- Add a link to the current MCP spec revision (2025-11-25) alongside the 2025-03-26
baseline on the protocol-version check.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Swetha Kumar <16600902+Swethakumar1@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Aaron Powell <me@aaron-powell.com>
* Add pester-should-migration skill (experimental)
Companion to the pester-migration skill. Focuses on the optional move from
the classic v5 `Should -Be` assertion syntax to the new v6 `Should-*`
assertions (e.g. `Should -Be` -> `Should-Be`). Includes a full
operator-by-operator mapping and the behavioral gotchas (negation as a
separate command, BeExactly -> Should-BeString -CaseSensitive, truthy/falsy
vs strict bool, BeNullOrEmpty split, collections, pipeline unwrapping).
Marked experimental/preview: verified against Pester 6.0.0-rc2; the new
Should-* assertions may still change before the 6.0 release.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Trim experimental status callout in pester-should-migration
Address review feedback: drop the time-based, human-oriented preview guidance (release-candidate timing, "as of mid-2026", release-notes link) from the status callout, keeping a lean experimental marker. The agent has no concept of time, so that prose is just extra tokens.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Awesome Copilot Community <awesome-copilot-community@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Aaron Powell <me@aaron-powell.com>
* Add pester-migration skill
A self-contained, experimental skill that helps upgrade PowerShell Pester
test suites across major versions (v3->v4, v4->v5, v5->v6). One router
SKILL.md plus per-jump references. The v5->v6 guidance tracks Pester 6,
which is still a release candidate.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Regenerate skills index for pester-migration
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review feedback on pester-migration skill
- Pin the stable-v5 install to -MaximumVersion 5.99.99 so it keeps
installing v5 after Pester 6 goes GA (SKILL.md).
- Make the baseline run command version-agnostic (bare Invoke-Pester)
and note that parameters differ across majors (SKILL.md).
- Replace the '->' mapping arrows inside powershell fences with
comment + valid replacement lines so snippets are copy/paste-safe
(SKILL.md v5->v6 cheat sheet, v5-to-v6.md, and v3-to-v4.md).
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review: scope -SkipPublisherCheck, trim status callout, order references
- Drop -SkipPublisherCheck from the default install commands; scope it to
Windows PowerShell 5.1 (installing over the OS's Microsoft-signed built-in
Pester 3) per pester.dev install docs, instead of an unconditional default.
- Trim the experimental status callout: keep the preview marker, drop the
date-based, human-oriented wording.
- Reorder the References table into version progression order (v3->v4, v4->v5,
v5->v6).
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: nohwnd <jakub@jares.cz>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Aaron Powell <me@aaron-powell.com>
* Add copilot-pr-autopilot skill
Skill that drives any GitHub pull request through repeated rounds of
Copilot Code Review until the agent has either resolved every thread
or explicitly escalated it to the human. Triggered via GraphQL (no
@copilot mention needed), triages every open thread with a fix /
decline / escalate rubric, replies and resolves each thread citing
the pushed SHA, then re-triggers until HEAD is reviewed with zero
threads awaiting the agent's reply.
Includes step scripts (01 request-review, 02 check-review-status,
03 list-open-threads, 08 reply-and-resolve, 10 cleanup-outdated),
shared library (_lib.ps1) with gh-CLI wrappers (Invoke-Gh,
Invoke-GhGraphQL, ConvertFrom-GhJson, Assert-GhReady), reply
templates, and reference docs for each step.
Repo-agnostic. Requires gh CLI on PATH and repo Triage/Write for
full autopilot; external PR authors get single-iteration mode with
manual re-trigger via the UI re-request button or a substantive
push.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review: split per-step references + add recap-gate circuit breaker
- Fix#1: give steps 1/7/10 their own reference files
(01-request-review.md, 07-commit-push.md, 10-cleanup.md); trim the
inline bodies out of orchestration.md so it stays cross-cutting only.
- Fix#3: add a recurring round-cap & recap gate to 09-convergence.md —
default STOP every 10th round, recap all prior rounds, detect drift
(out-of-scope / over-engineering / wrong-direction / belongs-in-separate-PR)
with CONTINUE / REVERT-AND-SHIP / HAND-OFF verdicts. Agent reasoning,
no new script.
- Surface the gate from SKILL.md and orchestration.md; regenerate
docs/README.skills.md. Markdown-only change; scripts unchanged.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* docs(copilot-pr-autopilot): surface recap gate in decision pseudo-code; clarify Copilot+human convergence and round definition
- Inject the round-cap recap gate into the '## Decision: loop back or exit'
pseudo-code else-branch so an agent following the code block (not just the
prose) runs the STOP-every-10th-round check before looping.
- Broaden the 'never terminal' paragraph: non-convergence is driven by a
Copilot finding OR a human review comment (this skill handles both); the
loop ends only when there are no new comments from either source AND every
open thread (Copilot or human) has an agent reply/escalation.
- Define a 'round' explicitly as one execution of step 1 (01-request-review),
i.e. one Copilot-review trigger — the cap counts review rounds, not tool
calls or fix edits.
Markdown-only; no script changes.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* copilot-pr-autopilot: make recap-gate round count deterministic
Add 09-review-round.ps1: counts Copilot Code Review submissions straight
from the PR's API history (full GraphQL pagination), so the recap-gate
trigger is a derived number, not a fallible agent mental tally. This
removes the exact failure mode the skill exists to survive — a count
drifting across a long run (the real 156-round case).
The script reports Round + RecapDue (Round % RecapInterval == 0) only;
it never stops the loop or picks the verdict. CONTINUE / REVERT-AND-SHIP
/ HAND-OFF stays agent reasoning. 09-convergence.md updated to reference
the deterministic count while preserving 'no script stops the loop' and
'non-convergence = Copilot finding OR human comment'.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Regenerate docs/README.skills.md for copilot-pr-autopilot (add 09-review-round.ps1)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* azure-devops-cli: handle long comments on Windows (#2061)
On Windows 'az' is a cmd.exe batch wrapper capped at ~8191 characters,
so a long --discussion / --description value silently truncates or
fails. Document three verified ways out so the coding agent doesn't
waste 3-5 turns falling back to raw token retrieval and REST:
1. azps.ps1 in PowerShell on Windows. Same Azure CLI, invoked through
the Python entry point with no cmd.exe length cap. Pair with
'Get-Content -Raw' so the body lives in a variable, not on the
command line.
2. Native --file-path flags where Azure CLI offers them. Applies to
'az devops wiki page create' and 'az devops wiki page update', both
documented with --encoding.
3. 'az devops invoke --in-file' as the universal escape hatch for
commands with no --file-path (work-item --discussion, PR
--description). Documented example posts to the work item
comments REST endpoint with api-version 7.0-preview.3.
The earlier draft suggested the Azure CLI '@<file>' convention as a
generic substitute for inline string args. The official docs only
document it for JSON parameters and the CLI source uses
'get_file_json' specifically, so the claim is removed and replaced
with an explicit warning not to rely on it for plain string args.
Files touched:
- skills/azure-devops-cli/SKILL.md: new 'Posting long comments on
Windows' section with shell-detection table and three verified
options.
- skills/azure-devops-cli/references/boards-and-iterations.md: short
pointer at each --discussion example back to SKILL.md, plus an
inline PowerShell snippet.
Closes#2061.
* azure-devops-cli: move long-comments guidance to reference file (#2061 review)
aaronpowell asked for the Windows long-comments section to live as a
reference file rather than inline in SKILL.md, so the token weight
isn't always loaded into the agent's context.
- Move the "Posting long comments on Windows" section to a new
references/long-comments-on-windows.md verbatim.
- Strip the section from SKILL.md (56 fewer lines in the always-loaded
surface).
- Add the new file to the Reference Files table in SKILL.md with a
one-line "when to read" hint covering --discussion, --description,
and --content failures on Windows.
- Update the two pointer comments in references/boards-and-iterations.md
to point at the new reference file instead of the SKILL.md section.
docs/README.skills.md regenerated by 'npm run build' to pick up the
new reference file in the skill's bundled assets column.
* Refresh 6 stale instruction files flagged in #2133
Targeted refresh:
- blazor: C# 13 to C# 14. Drop the Visual Studio Enterprise mandate so contributors on VS Code or Rider aren't blocked by a paid SKU. Swap VS-only profiling for dotnet-trace and dotnet-counters.
- copilot-thought-logging: narrow applyTo from '**' to '**/Copilot-Processing.md'. Replace 9 Windows backslash paths with POSIX './Copilot-Processing.md' so the workflow works on macOS and Linux.
- genaiscript: drop the "avoid exception handlers or error checking" line. Replace it with: handle errors at I/O and external API boundaries, let unexpected exceptions surface.
- memory-bank: add the required 'description' frontmatter field (was a validation failure). Narrow applyTo from '**' to 'memory-bank/**'. Add an opt-in note so contributors know auxiliary files land in the workspace root.
Minor modernization:
- azure-functions-typescript: Node.js v20 to v22 LTS.
- localization: relative '../../issues' disclaimer link to absolute https://github.com/github/awesome-copilot/issues so it resolves regardless of the localized doc's path.
docs/README.instructions.md regenerated by 'npm run build' to pick up the new memory-bank description.
* revert applyTo narrow on copilot-thought-logging (#2133 review)
aaronpowell flagged that narrowing applyTo from '**' to
'**/Copilot-Processing.md' inverts the instruction. The instruction
tells Copilot to CREATE Copilot-Processing.md when handling any user
request, so it must apply globally, not only when that file is
already open.
Restore applyTo to '**'. Keep the POSIX path fixes (backslash to
'./Copilot-Processing.md') and the other 5 file fixes in this PR
unchanged.
---------
Co-authored-by: Aaron Powell <me@aaron-powell.com>
Retarget instructions/astro.instructions.md to Astro 7 and
instructions/svelte.instructions.md to Svelte 5 / SvelteKit 2, fix
outdated APIs, add current stable feature guidance, and trim non-code
(setup, deployment, testing-workflow, recap) content so each file
focuses solely on code structure and best practices.
Regenerated docs/README.instructions.md via npm start.
Co-authored-by: GeekTrainer <GeekTrainer@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Adrien Clerbois
Burak Bayır
denis-a-evdokimov
Shubham Jiyani
skyestrela
Himanshu bamoria
NeilChen
Samuel Bushi
Jonas Tüchler
Ayan Gupta
Mostafa Amini
Baiqing Lyu
Simon He
Jenny Ferries
Suren K
Anton Standrik
Willie Yao
Suren K
Augustin Popa
prasadgd9022
Akash Dwivedi
Paul Delannoy
Gijs Reijn
samqbush
Swetha Kumar
Sunil Sattiraju
Muhammad Ubaid Raza
Finn Womack
Jakub Jareš
Gordon Lam
Komal Vardhan Lolugu
Christopher Harrison