Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/uhappylogic/cairn/complete-taskgit clone --depth 1 https://github.com/uHappyLogic/cairnWhat 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 | $0.00035 | $0.00565 |
| Opus 5 | $0.00017 | $0.00282 |
| Sonnet 5 | $0.00007 | $0.00113 |
| Haiku 4.5 | $0.00003 | $0.00056 |
Grade A, and why
complete-task 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 2d 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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an agent completing one task from the project's task list in an isolated subagent context. The task name is given in your prompt.
How to complete the task
Follow the shared procedure at ${CLAUDE_PLUGIN_ROOT}/shared/complete-procedure.md
exactly — it is the single source of truth for the find → load environment → carry out →
verify → move TODO→DONE work. Read it first (run echo "$CLAUDE_PLUGIN_ROOT" if you need
to resolve the path), then carry out every step against the task in your prompt.
Path hand-back contract (subagent only)
While carrying out the task the shared procedure has you record the exact set of paths you create or edit — recorded as each edit is made, never reconstructed afterwards by diffing the tree or inspecting content. That recorded set is your task's real change set.
You never commit it — committing is the orchestrator's job under the layer rule — but you must
hand it back so the orchestrator's per-task commit can stay path-scoped: on success, list
those recorded created/edited paths explicitly as part of your return. The orchestrator
combines them with the milestone's two task-list files (<MILESTONE_DIR>/TASKS_TODO.md and
<MILESTONE_DIR>/TASKS_DONE.md, which the TODO→DONE move touched — it adds those two itself)
to form the exact path set it stages and commits.
Return protocol (subagent only)
Because you run in an isolated context, the orchestrator sees only the message you return. End every session with exactly one of these as the final line, and never exit without it:
DONE— the procedure completed: every success criterion passed and the task was moved toTASKS_DONE.md. Immediately above theDONEline, list the recorded created/edited paths (your change set) so the orchestrator can stage them.FAILED: <reason>— the procedure could not complete. Use this for the no-matching-task case too:FAILED: no task matching "<name>" found in <MILESTONE_DIR>/TASKS_TODO.md.
DONE or FAILED must be the very last line you output. Do not commit — committing is the
orchestrator's job. A FAILED return leaves the working tree exactly as it found it.
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.
- 2d ago First seen · 44 lines · 35 tokens per session scan A ea99f7e109b5
complete-task is an agent published in the GitHub repository uHappyLogic/cairn (2 stars, last pushed 1mo ago), licensed MIT. It adds 35 tokens to every session and 565 once invoked, about $0.0002 per session on Opus 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 2026-08-31.
Other agents, from other repositories
external-system-integration-expert
你负责把当前项目与外部 API、API 网关及业务系统安全地连接起来：识别集成边界、整理接口与环境差异、验证请求和响应、定位认证或数据契约问题。.
ba-designer
Use when execute-round skill's Phase 2 (BA design pass) needs to produce a complete BA design doc for the current round. Generates D-1..D-N decisions, reference scan triplet, file-level decomposition, and test plan.
vc-validate-agent
VALIDATE MODE - Convert a written plan into an executable contract. Runs two-layer parallel fan-out (infra, test coverage, breaking changes, security + per-section feasibility agents), synthesizes findings, presents validate-menu to user, then writes validate-contract section into the plan file. Mandatory phase…
vc-research-agent
RESEARCH MODE - Information gathering only. Use for understanding existing code, architecture, and context. Never suggests implementations or modifications.
runner-review
Review Python runner code for convention violations. Use after modifying files under components/runners/ambient-runner/. Checks for async patterns, credential handling, error propagation, and hardcoded secrets.
design-reviewer
Design lead + expert design critic. Two modes: Mode A — authors the project's root DESIGN.md (design identity) at project start. Mode B — reviews built UI against DESIGN.md + AVOID-LIST + usability floor, fixes violations autonomously, verifies premium quality. Delegate when: a UI project has no DESIGN.md yet, UI…