Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/chris-dare-dev/agent-kitnpx agentmods add commands/chris-dare-dev/agent-kit/skill-to-commandWrote 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/commands/chris-dare-dev/agent-kit/skill-to-command)<a href="https://agentmods.dev/commands/chris-dare-dev/agent-kit/skill-to-command"><img src="https://agentmods.dev/badge/commands/chris-dare-dev/agent-kit/skill-to-command/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/commands/chris-dare-dev/agent-kit/skill-to-command"><img src="https://agentmods.dev/badge/commands/chris-dare-dev/agent-kit/skill-to-command.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 | $0.00145 | $0.05567 |
| Opus 5.5 | $0.00058 | $0.02227 |
| Sonnet 5.5 | $0.00029 | $0.01113 |
| Haiku 4.5 | $0.00015 | $0.00557 |
Grade A, and why
skill-to-command 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 — 364 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/skill-to-command — Skill→Command conversion orchestrator
Refactor an existing skill at data/skills/<name>/SKILL.md into the canonical slash-command + memory-bearing subagents pattern (the shape used by /milestone-pipeline, /spike, /roadmap).
Arguments: $ARGUMENTS — parse as <skill-name> [--skip-review] [--dry-run]
<skill-name>— required. Must exist atdata/skills/<name>/SKILL.md. If empty or absent, STOP and ask: "Which skill do you want to convert? (must exist atdata/skills/<name>/SKILL.md)"--skip-review— skip Turn 3 adversarial review. Use only for trivial conversions where the author is confident. Default is to always run the reviewer.--dry-run— author all files but do NOT commit. Useful for sanity-checking the conversion before staging.
When to use / When NOT to use
Convert with /skill-to-command |
Keep as a skill |
|---|---|
| Skill body has 3+ phases that each dispatch sub-work | Single-step skill (one prompt → one output) |
| Skill is >200 lines and growing | Skill is <100 lines and stable |
| Skill needs persistent memory across runs | Skill is purely stateless reference material |
| Skill currently uses TodoWrite to track multi-step state | Skill is a one-shot helper |
| Skill needs parallel agent dispatch in one of its phases | Skill is sequential by nature |
Key signal: if the skill body has Phase 1 / Phase 2 / Phase 3 / Phase 4 sections, parallel agent dispatch, or state.json, it is a candidate. If it has none of those, leave it as a skill — the conversion overhead is not worth it.
Inputs / outputs
| Input | Source |
|---|---|
| Skill body | data/skills/<name>/SKILL.md |
| Subdirectory references | data/skills/<name>/references/*.md (if present) |
| Subdirectory scripts | data/skills/<name>/scripts/* (if present) |
| Workspace CLAUDE.md | data/claude-md/workspace-root.md |
| Build pattern + 14 gotchas | data/references/skill-to-command-conversion-prompt.md |
| Output | Destination |
|---|---|
| Slash command orchestrator | data/commands/<name>.md |
| Phase subagents (N per phase) | data/agents/<name>-<role>.md |
Migrated references (flat-named, git mv) |
data/references/<name>-<file>.md |
Migrated scripts (flat-named, git mv) |
data/scripts/<name>-<file>.{sh,py} |
| Deprecated skill stub | data/skills/<name>-deprecated/SKILL.md |
| Build-time critique audit (UNTRACKED) | .claude/notes/<name>-build-critique.md |
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 First seen · 364 lines · 145 tokens per session scan A 8fdc5bba1093
skill-to-command is a command published in the GitHub repository chris-dare-dev/agent-kit (0 stars, last pushed 2mo ago), licensed MIT. It adds 145 tokens to every session and 5,567 once invoked, about $0.0006 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 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
speckit.bug.fix
Apply the remediation from a bug assessment and record what was changed.