APM is a package manager for AI-agent configuration, using a manifest to declare reusable instructions, skills, prompts, agents, hooks, plugins, and MCP components. Development teams use it to reproduce agent setups across projects and control their dependencies through policies and lockfiles, while the catalogue contains many of these agent components.
Borrowing it
Nothing to install: this file belongs to microsoft/apm. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/microsoft/apm/main/.agents/skills/autopilot-issue-triage-worker/SKILL.mdgit clone --depth 1 https://github.com/microsoft/apmWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/microsoft/apm/autopilot-issue-triage-worker)<a href="https://agentmods.dev/skills/microsoft/apm/autopilot-issue-triage-worker"><img src="https://agentmods.dev/badge/skills/microsoft/apm/autopilot-issue-triage-worker/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/microsoft/apm/autopilot-issue-triage-worker"><img src="https://agentmods.dev/badge/skills/microsoft/apm/autopilot-issue-triage-worker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00063 | $0.05886 |
| Opus 5.5 | $0.00025 | $0.02354 |
| Sonnet 5.5 | $0.00013 | $0.01177 |
| Haiku 4.5 | $0.00006 | $0.00589 |
Grade A, and why
autopilot-issue-triage-worker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 527 lines — stays where its author put it; the contents beside it link to each section on GitHub.
autopilot-issue-triage-worker -- Single-Issue Triage
The scheduler already selected this issue. Do not call
fetch_queue.py or triage_state.py. Do not sweep. Confirm the
single target number; missing number -> stop.
Advisory only. Read assets/label-contract.json before reasoning.
The caller supplies the repository's GOVERNANCE.md and CONTRIBUTING.md:
those human policies override persona instructions. If unavailable, do
not invent authority or review contacts; flag the missing context.
An accept recommendation is not acceptance. Labels and silence are not
approval. Only a responsible human maintainer approves scope, priority,
contributor invitations, review capacity, and release targeting.
No assignment needed. No ORIGIN or INTENT assigns contributors,
requests reviewers, or edits existing human assignment. Unattended automations (Agentic Workflows,
gh-aw, GitHub Actions, future scheduled runs) and actor-session
runners (direct user harness, Copilot App, Cloud Agent, Remote Agent)
are equally read-only for ownership.
This worker owns advisory writes even when summoned without a
scheduler: one comment plus triage/recommended and optional
classification from the contract allowlist. Never write human
decision labels. The scheduler must not comment or label.
Writes are optional via the activation card (write: on default).
Activation card
activation_card: on. Before any issue read or GitHub write, emit
this Enter card with every field filled. Missing field -> stop.
skill: autopilot-issue-triage-worker
skill_path: <resolved directory of this SKILL.md>
mode: run
subject: microsoft/apm#<issue-number>
path: triage
intent: advise one already-selected issue
origin: unattended | actor-session
write: on | off
json: off | on
debug: off | on
repo: microsoft/apm
issue: <positive integer>
invocation: agentic-workflow | actor-session
Rules:
writedefaults toonwhen the caller omitted it.write: offreturns the filled template only. Do not comment, add labels, or remove labels.write: onposts one GitHub comment viaautopilot-commentand processing / classification labels. Never human decision labels. Never assign. Do not callgh issue commentyourself.jsondefaults tooffwhen omitted or unknown. Omittedjsonis not a missing-field stop.json: off-> notriage-recommendationJSON receipt.json: on-> fill that JSON only as an internal payload (session file or parent Exit). Never post it on GitHub.debugdefaults tooffwhen omitted or unknown. Omitteddebugis not a missing-field stop.debug: off-> passpublic_bodyas the Suggested issue comment (unwrap the markdown fence) toautopilot-comment. Do not pass recommendation, classification, brief, next action, or persona details aspublic_body.debug: on-> passdebug_bodyas the filled template without the JSON fence.autopilot-commentprefixes[i] Skill debug is on.originfail-closed unknown ->unattended.- One issue. Do not nest a scheduler path.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · +30 lines 671433a866ce
- 1mo ago First seen · 497 lines · 63 tokens per session scan A a4fd4263638f
autopilot-issue-triage-worker is a skill published in the GitHub repository microsoft/apm (4,133 stars, last pushed yesterday), licensed MIT. It adds 63 tokens to every session and 5,886 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-17.
Other skills, from other repositories
bootstrap
Bootstrap a greenfield project with parent-owned interviews and bounded plan, revision, and apply actions.
blueprint
Guided parent-owned project charter and PRD interviews with durable artifact-only resume.
build-fast
Quick-iteration development for small/medium scope changes with persistent artifacts and optional review.
speedrun
Interactive speedrun loop for small, low-risk changes. Delegates each request to a general sub-agent. Redirects larger work to /build-fast or /build.
arcade-collab
Structured guidance for agents to read, classify, and act on user feedback (annotations and direct file edits) from the Arcade.
phase-plan
Decompose a completed PRD or oversized requirements artifact into durable delivery phases.