Equity-Research-Company: Command for Claude Code

.claude/commands/log-incident.md

log-incident is a command for Claude Code from pppop00/Equity-Research-Company. It costs 44 tokens per session (834 once invoked), scanned A, original, Apache-2.0.

A command that drafts a new incident record from the latest workspace run. An incident record documents what failed, why it failed, and the rule that should prevent the same problem later.

In plain words
What is it for?
Use it after a failed run to collect validator results, check the workspace outputs, and prepare a structured entry for INCIDENTS.md.
Why use it?
It turns a recent failure into a reviewable, permanent lesson without immediately changing the incident log. A person confirms the draft before it is added.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

This is pppop00/Equity-Research-Company's own configuration. It tells Claude Code how to work on Equity-Research-Company itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Equity-Research-Company configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 tools/io/log_incident.py --collect --workspace workspace/<latest_run>/ --description "$ARGUMENTS".

Reuse

Borrowing it

Nothing to install: this file belongs to pppop00/Equity-Research-Company. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/pppop00/Equity-Research-Company/main/.claude/commands/log-incident.md
Clone the repo
git clone --depth 1 https://github.com/pppop00/Equity-Research-Company

Made for: Claude Code.

Wrote 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.

agentmods badge for log-incident

README.md
[![agentmods](https://agentmods.dev/badge/commands/pppop00/equity-research-company/log-incident/github.svg)](https://agentmods.dev/commands/pppop00/equity-research-company/log-incident)
Your own site
<a href="https://agentmods.dev/commands/pppop00/equity-research-company/log-incident"><img src="https://agentmods.dev/badge/commands/pppop00/equity-research-company/log-incident/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.

agentmods 80×15 button for log-incident

Your own site · 80×15
<a href="https://agentmods.dev/commands/pppop00/equity-research-company/log-incident"><img src="https://agentmods.dev/badge/commands/pppop00/equity-research-company/log-incident.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 834 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00044 $0.00834
Opus 5 $0.00022 $0.00417
Sonnet 5 $0.00009 $0.00167
Haiku 4.5 $0.00004 $0.00083

Measured 12d ago against content hash 0432f70a2aaa, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

log-incident 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 12d 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.

.claude/commands/log-incident.md · 44 lines

How it starts

The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/log-incident

The user just observed a failure in the most recent run and wants to add a permanent rule to INCIDENTS.md so it cannot recur. This command implements the Curate beat of the Anamnesis Pattern (see references/anamnesis_pattern.md).

What you should do

  1. Read the user's $ARGUMENTS — a short description of what went wrong.

  2. Pick the latest workspace run — the most recently modified directory under workspace/{Company}_{Date}/. If none exists, ask the user to point you at the run dir.

  3. Run the collector to gather a structured digest of that run:

    python3 tools/io/log_incident.py --collect --workspace workspace/<latest_run>/ --description "$ARGUMENTS"
    

    This prints a JSON digest of the run's validator outputs, structure-conformance, red-team outputs (if present), and the resolved P0 gate values. It does not write to INCIDENTS.md.

  4. Read INCIDENTS.md to see the existing entry format and the next available I-NNN id.

  5. Draft a candidate entry matching the existing format (Phase, What happened, Root cause, Rule (load-bearing), Detection, Related contract). Use concrete paths and values from the digest, not generic prose.

  6. Show the candidate entry to the user as a proposed INCIDENTS.md append — do not write it yet.

  7. Wait for explicit user confirmation. Acceptable confirmations: yes, confirm, append, ok 写入, OK. Anything else (including silence) means do not append.

  8. On confirmation, append the entry to INCIDENTS.md (do not rewrite or reorder existing entries; append-only). Print the path and the appended id back to the user.

Hard rules

  • Append only. Never delete, reorder, or rewrite existing INCIDENTS.md entries. If a past entry is now incorrect, supersede it with a new entry that links back via Related contract:.
  • Human gate is non-negotiable. Do not append without an explicit user confirmation. Auto-mode does not override this.
  • No PII. If the digest contains the user's email (e.g. surfaced from a User-Agent string in data_source fields), strip it before quoting in the entry.
  • Specific, not generic. Cite the actual run dir path, the actual workspace JSON paths, the actual values that were wrong. Generic entries don't earn a permanent slot in the system prompt.
  • One rule per entry. If the user describes two distinct failures, draft two entries and confirm separately.

Read the full file on GitHub · 44 lines

Changes

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.

  1. 12d ago First seen · 44 lines · 44 tokens per session scan A 0432f70a2aaa

Subscribe to this mod's changes

log-incident is a command published in the GitHub repository pppop00/Equity-Research-Company (8 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 44 tokens to every session and 834 once invoked, about $0.0002 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.