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/jackguo709/devloop/grownpx skills add jackguo709/devloop --skill growgit clone --depth 1 https://github.com/jackguo709/devloopWhat 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 | $0.00016 | $0.01885 |
| Opus 5 | $0.00008 | $0.00942 |
| Sonnet 5 | $0.00003 | $0.00377 |
| Haiku 4.5 | $0.00002 | $0.00188 |
Grade A, and why
grow scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context
DEVLOOP_DIR="$HOME/.devloop"
PROFILE="$DEVLOOP_DIR/profile.md"
mkdir -p "$DEVLOOP_DIR/observations"
[ -f "$PROFILE" ] && echo "PROFILE_EXISTS=true" || echo "PROFILE_EXISTS=false"
echo "DEVLOOP_DIR=$DEVLOOP_DIR"
if [ -f "$HOME/.devloop/scripts/check-grow-ready.sh" ]; then
bash "$HOME/.devloop/scripts/check-grow-ready.sh" --context 2>/dev/null
else
echo "GROW_AVAILABLE=true"
fi
Routing
If PROFILE_EXISTS=false, read onboarding.md and follow its instructions.
If PROFILE_EXISTS=true:
- No arguments or empty
$ARGUMENTS:- If
GROW_AVAILABLE=false: respond with the value ofCOOLDOWN_MSGas plain text and stop. Do not read any other files. - Otherwise → read run.md and follow its instructions
- If
$ARGUMENTS= "me" → read me.md and follow its instructions$ARGUMENTS= "history" → read history.md and follow its instructions
Find Content
ONLY use content from sources listed in sources.md. Run this entire section silently.
Step 1: Target
Read profile.md, observations, and history.jsonl. Identify 3-5 growth opportunities, ranked with reasoning. Don't search yet. For each:
- Which concept(s) it addresses
- Impact: A 0-1 concept or recurring cross-domain pattern from Deep Patterns? Does fixing this unlock progress in other domains? A tool that eliminates their current approach? Generally: product direction > tools/workflow > approach > skill gaps, but not rigid. Mindset-level feedback is the most personal and requires established trust.
- Readiness: Is the user bumping into this right now, or is it important in the abstract? A user mid-build may need building advice. A user who just shipped may need distribution advice.
- Pacing: Repeat topics from history.jsonl only if strategic. Think twice before pushing topics where past recommendations were ignored.
Step 2: Explore
Two parallel Sonnet subagents. They explore wide and return a menu of candidates, not a committed topic:
What ships with it
16 files 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.
- history.md 132 B
- me.md 108 B
- onboarding.md 7.7 KB
- references/domains/agent-fluency.md 8.7 KB
- references/domains/business.md 6.5 KB
- references/domains/directing-ai.md 8.1 KB
- references/domains/distribution.md 7.3 KB
- references/domains/product-taste.md 7.0 KB
- references/domains/resilience.md 6.4 KB
- references/domains/verification.md 9.8 KB
- references/permissions.json 1.7 KB
- references/sources.md 1.7 KB
- references/visual-identity.md 2.9 KB
- run.md 4.9 KB
- scripts/check-grow-ready.sh 1.9 KB runs code
- scripts/detect-stack.sh 10 KB runs code
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 · 209 lines · 16 tokens per session scan A f4550e46307f
grow is a skill published in the GitHub repository jackguo709/devloop (2 stars, last pushed 4mo ago), licensed MIT. It adds 16 tokens to every session and 1,885 once invoked, about $0.0001 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
skill-tutor
Use this skill whenever the user wants to learn how to manually do what an existing Claude skill does, or when they want to understand the underlying process of any capability Claude performs automatically. Triggers include: "how can I learn to do X?", "what do I need to know to do X without Claude?", "teach me what…
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…