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 skills/crypticswarm/swarmforge/skill-reflectionnpx skills add CrypticSwarm/Swarmforge --skill skill-reflectiongit clone --depth 1 https://github.com/CrypticSwarm/SwarmforgeWhat 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 | $0.00043 | $0.00783 |
| Opus 5 | $0.00022 | $0.00392 |
| Sonnet 5 | $0.00009 | $0.00157 |
| Haiku 4.5 | $0.00004 | $0.00078 |
Grade A, and why
skill-reflection 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 yesterday.
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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Reflection
Use this skill to produce a high-signal retrospective on the current conversation focused on skills: what was used, what was missing, and what would make future runs faster and safer.
Output Contract
Return a short, structured reflection. Prefer bullets over prose.
Include these sections when applicable:
- Conversation recap (1–3 bullets): what the user wanted and what happened.
- Skills used: list explicitly activated skills and any implicitly-triggered ones (name + evidence).
- Skill improvements: actionable recommendations to update existing skills.
- New skills to consider: candidates for repeatable workflows discovered in this conversation.
- Commands to consider (optional): whether a slash command would reduce friction.
How to Identify Skills Used
- Explicit usage: look for direct skill activation (for example,
skill: <name>tool calls) and commands that instruct activation. - Implicit usage: infer from the agent’s behavior and/or command prompts (for example, a commit message workflow likely used
commit-messages). - If uncertain, say so explicitly (for example, “Likely used
general-software-engineering(pattern match), but not explicitly activated”).
How to Recommend Skill Improvements (Keep It General)
Your goal is to improve the skill package, not to solve the one-off task.
When suggesting a change, include:
- Problem pattern: what repeated friction or failure mode appeared.
- Proposed change: what to add/remove/clarify in the skill (description triggers, workflow steps, safety constraints, validation commands, templates).
- Why it helps: how it reduces ambiguity, prevents mistakes, or saves iterations.
Guardrails:
- Avoid recommendations that are overly specific to a single repo, file path, or one-off request.
- It’s fine to cite the conversation as an example, but phrase the change as a general technique.
- Prefer small edits that preserve the skill’s intent and minimize token bloat.
- If the gap is better solved by a command (argument validation + small context blocks), say so.
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.
- yesterday First seen · 79 lines · 43 tokens per session scan A aab3dbaaa1f4
skill-reflection is a skill published in the GitHub repository CrypticSwarm/Swarmforge (2 stars, last pushed 2d ago), licensed MIT. It adds 43 tokens to every session and 783 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…