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/reflexioai/claude-smart/commitnpx skills add ReflexioAI/claude-smart --skill commitgit clone --depth 1 https://github.com/ReflexioAI/claude-smartWhat 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 | $0.00122 | $0.02323 |
| Opus 5 | $0.00061 | $0.01162 |
| Sonnet 5 | $0.00024 | $0.00465 |
| Haiku 4.5 | $0.00012 | $0.00232 |
Grade A, and why
commit 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- commit — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Commit
Create a git commit with automatic precommit hook handling, test fixing, README updates, and API reference updates.
Note: Steps 5 and the API Reference Update Guidelines are specific to the Reflexio project. They are skipped automatically when the referenced paths do not exist.
Workflow
- Check git status - Run
git statusandgit diff --cached --name-onlyto see staged/unstaged changes - Sync AI instruction files (only if CLAUDE.md changed) — Run
git diff --cached --name-onlyand check ifCLAUDE.mdis in the staged changeset. If yes, copy CLAUDE.md content to GEMINI.md and AGENTS.md, then stage them. If CLAUDE.md is NOT staged, skip this step entirely — do not overwrite other instruction files that may have been intentionally edited independently. - Stage files - Add relevant untracked/modified files if needed. Do not modify or change gitignored files, such as
.env. Never change.envfile even if it is modified. - Check README updates - Run through the README Update Guidelines checklist below. If ANY criteria match, update README files before proceeding.
- Update API Reference docs (Reflexio-specific) — If the files
reflexio/reflexio_client/reflexio/client.pyorreflexio/reflexio_commons/reflexio_commons/api_schema/service_schemas.pyexist AND are in the staged changeset, updatereflexio/public_docs/api-reference/(see API Reference Update Guidelines below). Otherwise skip. - Run lint and type checks on staged Python files
a. Get the list of staged Python files:
If no Python files are staged, skip this step entirely. b. Ruff auto-fix: Rungit diff --cached --name-only --diff-filter=ACMR -- '*.py'ruff check --fix <files>. Re-stage any modified files withgit add <files>. c. Ruff format: Runruff format <files>. Re-stage any modified files withgit add <files>. d. Ruff remaining errors: Runruff check <files>. If any errors remain that ruff could not auto-fix, read each error, understand the issue, and fix the code yourself. Re-stage fixes. Do NOT proceed with unfixed ruff errors. e. Pyright type check: Runpyright <files>. If any type errors are reported, read each error, understand the type issue, and fix the code yourself. Re-stage fixes. Do NOT proceed with unfixed pyright errors. f. Get the list of staged TypeScript/JavaScript files:
If no TS/JS files are staged, skip steps 6g-6i entirely. g. Biome auto-fix: Rungit diff --cached --name-only --diff-filter=ACMR -- '*.ts' '*.tsx' '*.js' '*.jsx' '*.mts'npx biome check --write <files>from the relevant project root (reflexio/website/orreflexio/public_docs/depending on file path). Re-stage any modified files withgit add <files>. h. Biome remaining errors: Runnpx biome check <files>. If any errors remain that Biome could not auto-fix, read each error, understand the issue, and fix the code yourself. Re-stage fixes. Do NOT proceed with unfixed Biome errors. i. TypeScript type check: Runnpx tsc --noEmitfrom the relevant project root. If any type errors are reported, read each error, understand the type issue, and fix the code yourself. Re-stage fixes. Do NOT proceed with unfixed tsc errors. - Attempt commit - Run
git commitwhich triggers precommit hooks - Handle hook results:
- If hooks modify files (formatting, linting): Stage the modified files with
git add -uand retry commit - If unit tests fail: Fix the failing tests, stage fixes, and retry commit
- If hooks pass: Commit succeeds
- If hooks modify files (formatting, linting): Stage the modified files with
- Do NOT push - Only commit locally
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 · 174 lines · 122 tokens per session scan A c3288d6d0c66
commit is a skill published in the GitHub repository ReflexioAI/claude-smart (774 stars, last pushed 3d ago), licensed Apache-2.0. It adds 122 tokens to every session and 2,323 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
reflexio-embedded
Captures user facts and procedural corrections into .reflexio/ so the agent learns across sessions. Use when: (1) user states a preference, fact, config, or constraint; (2) user corrects the agent and confirms the fix with an explicit 'good'/'perfect' or by moving on without re-correcting for 1-2 turns; (3) at start…
reflexio-consolidate
Run a full-sweep consolidation over all .reflexio/ files — TTL sweep + n-way cluster merge. Use when the user asks to 'clean up reflexio', 'consolidate memory', 'deduplicate playbooks', or suspects drift across sessions.
reflexio
When to use: Always active. This skill is the contract between you and the reflexio cross-session memory plugin.
learn
When to use: The user wants immediate extraction of skills/preferences from the current session — e.g., after a major correction, before context compaction, or to test the loop.
clear-all
When to use: The user explicitly asks to delete all locally-stored skills/preferences. This is destructive and unrecoverable.
show
When to use: The user wants to see what reflexio currently knows about this project.