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/agentculture/fleet-cli/thinknpx skills add agentculture/fleet-cli --skill thinkgit clone --depth 1 https://github.com/agentculture/fleet-cliWhat 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 | $0.00191 | $0.02732 |
| Opus 5 | $0.00096 | $0.01366 |
| Sonnet 5 | $0.00038 | $0.00546 |
| Haiku 4.5 | $0.00019 | $0.00273 |
Grade A, and why
think 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.
This is a copy
100% identical to think — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
think — work an idea backwards into a buildable spec
The skill is named think; the product/CLI it drives is devague. (The
forward leg — turning a converged spec into a plan — is the sibling
/spec-to-plan skill, which drives devague plan.)
think turns a vague feature idea into a buildable spec by working
backwards: you start from the announcement you'd make if it had already
shipped, then build an Announcement Frame by capturing claims, pressure
-testing them, parking what's still genuinely unknown, and only exporting once
the frame converges.
The CLI is deterministic and move-driven — it is not a wizard. There is no
fixed sequence of prompts. You (the agent) choose the next move; the CLI just
tracks state and tells you what's still missing. Run devague learn for the
canonical ten-stage arc and devague explain <move> for any single move.
This skill is the operator: a portable wrapper that resolves the CLI and
forwards every move verbatim — including status, the read-only verb that reads
the convergence gate and tells you the recommended next move.
How to run
The entry point is scripts/think.sh. Invoke it from the repository you are
speccing (frames persist under .devague/ in the current directory):
bash .claude/skills/think/scripts/think.sh <move> [args...]
bash .claude/skills/think/scripts/think.sh status
It resolves the CLI portably — an installed devague on PATH (the normal
case), falling back to uv run devague when you are inside the devague checkout.
If neither resolves it prints an install hint (uv tool install devague). Every
move — including status — is forwarded verbatim, so you can equally call the
CLI directly (devague <move> …) when it is installed; the wrapper exists only
for portable resolution.
Moves
| Move | What it does |
|---|---|
new "<announcement>" |
Start a frame from the announcement (the first move). Seeds an auto-confirmed announcement claim. |
capture --kind <kind> "<text>" |
Record + classify a claim. --origin llm lands it as proposed. |
interrogate <id> --honesty "…" |
Attach an honesty condition (what must be true). Also --hard-question, --risk, --contradicts, --blocking. |
confirm <id> [<id>…] / reject <id> [<id>…] |
Resolve one or more claims (c*) / honesty conditions (h*) in one transactional call. User-only decision. Also confirm --from-review <file> to apply an edited review artifact. |
review |
List every proposed (unconfirmed) claim + honesty condition with ids (--json too); writes a non-authoritative artifact to .devague/reviews/<slug>.md. Un-gated; never mutates. |
question "<text>" |
Record / list / --resolve a pending user decision as durable working state in .devague/questions/<slug>.md. |
park "<text>" --kind <kind> |
Move uncertainty into first-class open vagueness instead of forcing an answer. |
converge |
Evaluate the gate; list remaining gaps. |
export |
Write the buildable spec to docs/specs/ — only after converge passes. |
status |
Read-only: where the frame stands + the recommended next move (--json too). |
show / list |
Render a frame / list frames (--json for raw state). |
learn / explain <move> |
Teach the method / explain one move. |
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.
- yesterday First seen · 202 lines · 191 tokens per session scan A 3aced3418555
think is a skill published in the GitHub repository agentculture/fleet-cli (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 191 tokens to every session and 2,732 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to think, differing in 0 lines, and is treated as a copy.
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…