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 commands/alifanov/darkflow/fix-ci-issuegit clone --depth 1 https://github.com/alifanov/darkflowWrote 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/alifanov/darkflow/fix-ci-issue)<a href="https://agentmods.dev/commands/alifanov/darkflow/fix-ci-issue"><img src="https://agentmods.dev/badge/commands/alifanov/darkflow/fix-ci-issue.svg" alt="Measured on agentmods" 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 | $0.00029 | $0.01335 |
| Opus 5 | $0.00015 | $0.00668 |
| Sonnet 5 | $0.00006 | $0.00267 |
| Haiku 4.5 | $0.00003 | $0.00134 |
Grade A, and why
fix-ci-issue scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pick up one CI-failure task (source:ci), fix the failing lint/test, and push — with bounded retries. The same task is retried up to 3 times; only after the third failed attempt is it handed to a human. This prevents an endless red-CI → task → fix → red-CI loop.
The task itself is not closed here — the darkflow-ci-gate workflow closes it automatically once CI goes green again (close-on-green, via the Dark Flow task store). That is what makes the retry counter reliable: one open task per failing branch, attempts accumulate on it as comments.
Step 1 — Read project config
Load the project config (contract in .darkflow.d/claude.md → Project config). Uses: branch, mergeStrategy, language.
Step 2 — Pick the next CI task
Pick exactly one open task with source:ci and status:approved (a task already parked for a human carries status:needs-human, so it is out of this list by construction). Take the oldest (smallest number) — one open task per failing branch, FIFO:
n=$(~/.darkflow/df task list --source ci --status approved --state open | jq -r '
[ .[].number ] | min // empty')
If $n is empty, stop — nothing to do this run.
Step 3 — Count prior attempts (the retry gate)
Each attempt this command makes leaves a marker comment containing <!-- darkflow:ci-attempt -->. Count them from the task's comments (df task view returns the full comment array, or null when there are none):
attempts=$(~/.darkflow/df task view "$n" | jq '
[ (.comments // [])[] | select(.body | contains("<!-- darkflow:ci-attempt -->")) ] | length')
MAX_ATTEMPTS = 3.- If
attempts >= 3→ escalate to a human and STOP (see Step 6). Do not attempt another fix. - Otherwise this run is attempt
attempts + 1of 3 — continue to Step 4.
Step 4 — Reproduce and fix
Read the task with ~/.darkflow/df task view "$n" — the body lists the failing checks (e.g. lint, test) and the captured error output, plus the CI run link and commit SHA.
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.
- 5d ago First seen · 79 lines · 29 tokens per session scan A 57e8d5a359cd
fix-ci-issue is a command published in the GitHub repository alifanov/darkflow (2 stars, last pushed 6d ago), licensed MIT. It adds 29 tokens to every session and 1,335 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 commands, from other repositories
flow-nexus-workflow
Event-driven workflow automation with message queues.
release-manager
Automated release coordination and deployment with ruv-swarm orchestration for seamless version management, testing, and deployment across multiple packages.
release-swarm
Orchestrate complex software releases using AI swarms that handle everything from changelog generation to multi-platform deployment.
workflow-automation
Integrate AI swarms with GitHub Actions to create intelligent, self-organizing CI/CD pipelines that adapt to your codebase.
github-init
Initialize ClaudeForge CI/CD system.
watch
Create a FlowTrigger from a template AND generate a /loop prompt file the user invokes manually. Subcommands: pr , ci, issue , branch. Flow CANNOT invoke /loop from plugin code; this command prepares the prompt and tells the user how to start the loop themselves.