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 skills add OutlineDriven/outline-driven-development --skill cycle-memogit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/cycle-memo)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/cycle-memo"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/cycle-memo/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/outlinedriven/outline-driven-development/cycle-memo"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/cycle-memo.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00046 | $0.01141 |
| Opus 5 | $0.00023 | $0.00571 |
| Sonnet 5 | $0.00009 | $0.00228 |
| Haiku 4.5 | $0.00005 | $0.00114 |
Grade A, and why
cycle-memo 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cycle memo
Contract
| Field | Bound contract |
|---|---|
| Trigger | After a build, QA pass, demo, user complaint, or abandoned attempt when the useful output is what the next pass must learn rather than the code itself. |
| Authority | Reversible local: writes or refreshes only local docs under the project; rollback is restoring or deleting those docs. No VCS, credential, published, deployed, or remote mutation. |
| Side effect | Writes or refreshes local docs for the cycle: a memo for lessons and a plan for next-cycle contracts, gates, and vocabulary. |
| Done | Every lesson traces to observed cycle evidence; every anti-pattern names a concrete failure mode and the gate that catches it; no lesson is a single reported case in disguise (each covers a class with specifics as evidence); a fresh agent can tell what to preserve, discard, and test first without reading the whole old session. |
Inputs
Required: the original objective for the cycle and the final artifact it produced. Without both, lessons cannot be traced to evidence and the skill stops blocked.
Optional but cited when present: QA evidence, user complaints, screenshots, transcripts, diffs, and local planning notes.
Procedure
- Bound scope before mutation: extract lessons, not a changelog. File lists and effort summaries are not lessons. Do not defend the artifact; if it missed the product, say what missed. Done when: the scope is bounded to lesson extraction, not changelog or defense.
- Gather cycle inputs: the original objective, the final artifact, QA evidence, user complaints, and any local planning notes. If the objective or final artifact is unavailable, stop and report blocked. Done when: the objective and final artifact are gathered, or the run stops blocked.
- Separate working assets from misleading progress. List contracts, schemas, tests, services, vocabulary, and examples that earned reuse; set aside UI, panels, scaffolds, or abstractions that only looked productive. Preserve negative corpus: failed paths are training data, not deletions. Done when: working assets and misleading progress are separated, with failed paths preserved as negative corpus.
- Generalize before recording. When feedback arrives as many specific complaints, cluster them upward into the shared root gap before writing anything. Record the root and its pattern family as the lesson; keep the specifics under it as evidence, not as lessons. A lesson that only fixes the reported case keeps the system human-in-the-loop: the same class returns anywhere a person did not point. Escalate every specific to the pattern that catches its whole class; the reporter's exact words are evidence, not the gate. Done when: every specific complaint is escalated to its pattern-level lesson with specifics kept as evidence.
- Name each failure as an anti-pattern naming a concrete failure mode, not a mood. Done when: every failure is named as a concrete failure mode, not a mood.
- Convert repeated or high-impact failures into quality gates for the next pass. Prefer hard gates over advice. Done when: repeated or high-impact failures have hard gates, not advice.
- Convert vague user direction into architecture vocabulary a fresh agent can use. Done when: vague direction is converted to architecture vocabulary.
- Write or refresh local docs for the cycle: a memo for lessons and a plan for next-cycle contracts, gates, and vocabulary. Cite evidence from the cycle for every lesson: objective, file facts, QA output, screenshots, transcripts, diffs, or user feedback. Keep provenance local; shipped artifacts must not describe their development history. If the next agent cannot act on a lesson, it is not a lesson yet. Done when: the memo and next-cycle plan are written with every lesson citing cycle evidence.
- Verify a from-scratch agent could avoid the same failure from those docs alone: every lesson traces to observed cycle evidence; every anti-pattern names a concrete failure mode and the gate that catches it; no lesson is a single reported case in disguise; a fresh agent can tell what to preserve, discard, and test first without reading the whole old session. Done when: a from-scratch agent could avoid the same failure from the docs alone.
What ships with it
1 file 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.
- 2d ago Changed · -29 tokens per session 184db79f0e30
- 4d ago First seen · 45 lines · 75 tokens per session scan A 6efef155e2cd
cycle-memo is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed 3d ago), licensed Apache-2.0. It adds 46 tokens to every session and 1,141 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 2026-09-03.
Other skills, from other repositories
audit-project
Run an iterative multi-agent code audit until critical and high findings are resolved. Use when the user says "audit my code", "find all the bugs", "deep code audit", "iterative review", or "review until clean".
duet
Use when the user invokes /duet, says "pair with me", or faces aesthetic, architectural, or irreversible decisions.
goal-prompt-drafting
Use when asked to draft copy-ready /goal objectives for long-running agents; returns one normalized one-line objective with measurable end state, grounded proof, easy-out invariants, a stop clause, and a Missing list. Not for source or remote-system changes.
handoff-prompt
Use when the user asks for a handoff, delegation, or clipboard-ready prompt for another agent: a standalone path-free prompt copied to the clipboard, confirmed by title. Not for session-snapshot briefs — use handoff; never remote, credential, publish, deploy, or irreversible.
publish-branch
Use when asked to publish the checked-out branch: commit and push it on whatever branch it is, the default branch included. Not for creating branches, PRs, force pushes, or pushing any other branch; when the request excludes the default branch, use commit-push-current.
drill
Use when a concept needs practising rather than explaining: run a scaffolded exercise from worked example to independent problem, quiz the learner, run spaced recall over what they cleared, or probe for the gaps blocking what they want next. For explanation, use explain-concept; for an end-to-end build, use capstone.