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 skills add Bidiche49/claude-conf --skill backlog-impgit clone --depth 1 https://github.com/Bidiche49/claude-confWrote 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/skills/bidiche49/claude-conf/backlog-imp)<a href="https://agentmods.dev/skills/bidiche49/claude-conf/backlog-imp"><img src="https://agentmods.dev/badge/skills/bidiche49/claude-conf/backlog-imp/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/skills/bidiche49/claude-conf/backlog-imp"><img src="https://agentmods.dev/badge/skills/bidiche49/claude-conf/backlog-imp.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 | $0.00013 | $0.00697 |
| Opus 5.5 | $0.00005 | $0.00279 |
| Sonnet 5.5 | $0.00003 | $0.00139 |
| Haiku 4.5 | $0.00001 | $0.00070 |
Grade A, and why
backlog-imp 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create a new IMPROVEMENT ticket in the backlog.
Input
$ARGUMENTS
Steps
-
Check that
BACKLOG/exists. If not, tell the user to run/backlog-initfirst and stop. -
Calculate the next ID by scanning files (NEVER read INDEX.md for the ID):
- List all files in
BACKLOG/IMPROVEMENTS/PENDING/andBACKLOG/IMPROVEMENTS/DONE/ - Extract all numbers from
IMP-XXX.mdfilenames - Next ID = max number found + 1 (or 001 if no files exist)
- Format:
IMP-XXX(zero-padded to 3 digits)
- List all files in
-
Analyze the description to determine:
- Priorite (Critique / Haute / Moyenne / Basse)
- Complexite (XS / S / M / L / XL)
- Tags (free-form, relevant to the improvement)
-
Check for external documentation needs: If the improvement involves a third-party library, API, or service (e.g. migrating to a new SDK version, adopting a framework pattern), note in the ticket that the relevant documentation should be consulted before starting. Add a line in the description:
📖 Consulter: [doc URL or doc name]. -
Identify affected files: Based on the description and your knowledge of the codebase, determine which files will be modified. List actual paths. For uncertain candidates, add
(?). NEVER leave(A determiner). -
Write specific acceptance criteria: Describe the measurable improvement — what changes concretely? Performance gain, reduced complexity, better DX? Each criterion must be verifiable. Example: "Le hook s'execute en < 100ms au lieu de 500ms" instead of "C'est plus rapide".
-
Write actionable test steps: List the exact commands or scenarios to validate the improvement. Reference existing test infrastructure when applicable.
-
Create
BACKLOG/IMPROVEMENTS/PENDING/IMP-XXX.mdusing this format:# IMP-XXX: [Short title derived from description] **Type:** Improvement **Statut:** A faire **Priorite:** [determined] **Complexite:** [estimated] **Tags:** [relevant] **Depends on:** none **Blocked by:** — **Date creation:** [today YYYY-MM-DD] --- ## Description [User's description, enriched with current state vs desired state] ## Fichiers concernes - `path/to/file.ext` — [what changes in this file] ## Criteres d'acceptation - [ ] [Specific measurable improvement] ## Tests de validation - [ ] [Exact command or verification scenario]
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 · 72 lines · 13 tokens per session scan A e2e89cd501ec
backlog-imp is a skill published in the GitHub repository Bidiche49/claude-conf (2 stars, last pushed 5mo ago), licensed MIT. It adds 13 tokens to every session and 697 once invoked, about $0.0001 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 skills, from other repositories
{{BUCKET}}-management
Skill "{{BUCKET}}-management" from damusix/atomic-claude, covering purpose, doc shape and frontmatter contract.
config-manager
Skill that manages configuration.
code-review-helper
A safe skill that helps with code review tasks.
system-admin-helper
Skill with dangerous system commands.
vc:team
Orchestrate Agent Teams for parallel multi-session collaboration. Use for research, implementation, review, and debug workflows requiring independent teammates.
tour-de-controle
Orchestre une tâche complexe en trois temps avec Claude Code — PLANIFIER avec le modèle le plus puissant (Fable 5 / Opus), DÉLÉGUER l'exécution à des sous-agents calibrés lot par lot (Haiku / Sonnet / Opus selon la complexité) lancés en parallèle, puis VÉRIFIER chaque livrable de façon adversariale avec le modèle…