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 saski/arnesto --skill source-command-bug-fixing-agentgit clone --depth 1 https://github.com/saski/arnestoWrote 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/saski/arnesto/source-command-bug-fixing-agent)<a href="https://agentmods.dev/skills/saski/arnesto/source-command-bug-fixing-agent"><img src="https://agentmods.dev/badge/skills/saski/arnesto/source-command-bug-fixing-agent.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 | $0.00072 | $0.00773 |
| Opus 5 | $0.00036 | $0.00387 |
| Sonnet 5 | $0.00014 | $0.00155 |
| Haiku 4.5 | $0.00007 | $0.00077 |
Grade A, and why
source-command-bug-fixing-agent 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 3d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Company Environment Bug Fixing
Use this skill as the company-context overlay for a disciplined bug-fixing
workflow. Load and follow the diagnose skill for the core
reproduce-minimise-hypothesise-instrument-fix loop.
Inputs
Collect what is available, without making every item a prerequisite:
- Ticket, incident, or bug report URL and its relevant fields.
- Observed behavior, expected behavior, environment, timestamps, and error signatures.
- Candidate service, repository, workspace, package, or deployment.
- Available access to internal documentation, logs, issue trackers, source control, and CI/CD evidence.
Ask one focused question only when a missing input makes the next action unsafe or would materially change scope. Do not fabricate access, connector results, ticket fields, documentation, owners, or production state. When company access is unavailable, work only from artifacts the user provides and label the gaps.
Safety Boundary
- Never expose credentials, tokens, customer data, or restricted internal material in prompts, logs, patches, or external services.
- Pause for authentication, OTPs, secrets, production mutation, deployment, rollback execution, external messages, and ticket or pull-request updates unless the user has explicitly authorized that action.
- Treat incident mitigation and permanent correction as separate decisions.
- Preserve unrelated work and inspect repository instructions before editing.
- Do not send private company code or data through an unapproved model route.
Evidence Workflow
- Establish the bug contract from the ticket or report: observable failure, expected outcome, affected environment, scope, severity, and reproducibility.
- Search only available company sources for matching component names, error signatures, prior incidents, runbooks, architecture decisions, and recent changes. Record source strength and unresolved contradictions.
- Map the failure to code. Inspect stack traces, workspace metadata,
organization-prefixed dependencies such as
@work/*, service manifests, recent commits, andCODEOWNERSwhen present. Resolve packages from actual repository configuration rather than assumed paths. - Reproduce the failure at the narrowest public interface. Add one failing behavior-level test when practical.
- Rank a small set of falsifiable hypotheses and gather evidence that can disprove each one. Distinguish confirmed root cause from plausible causes.
- Implement the smallest reversible fix, then run the focused test and the repository's canonical validation. Verify relevant security or production risk only when the evidence makes it material.
- Define rollback and operational verification before recommending rollout.
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.
- 3d ago First seen · 76 lines · 72 tokens per session scan A 3e630a6500d8
source-command-bug-fixing-agent is a skill published in the GitHub repository saski/arnesto (5 stars, last pushed today), licensed Unlicense. It adds 72 tokens to every session and 773 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
runtime-evidence-and-tracing
Use when connecting observed behavior, logs, metrics, request IDs, run IDs, screenshots, traces, external dependency results, or artifacts into a runtime evidence loop.
html-ppt-zhangzara-block-frame
Rescuing a messy startup deck into a board-grade system — the diagnosis, the page grammar, and the rebuilt proof pages. Built as a decision-grade design craft deck for founders, exec presenter.
build-test
Run the project's build / typecheck / lint / test commands and emit the build.passing + tests.passing signals devloop convergence reads.
doca-flow
Build and debug DOCA Flow applications on supported NVIDIA NICs/DPUs: define match/action pipes, initialize ports and representors, choose forwarding targets, validate pipes before hardware programming, read counters, match the Flow version to the installed DOCA release, and diagnose Flow API errors. Trigger on DOCA…
mcore-create-issue
Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.
doca-dpa-hl-tracer
Use this skill when the user runs docadpahltracer to capture/decode DPA-side traces at the programming-events layer (kernel entry/exit, sync points, comm primitive calls, RDMA WR submission, completion drain) — picking TRACE vs CRIT, tuning the JSON config (file-size limits + filesizelimitpolicy, thread…