Skip to content
Agentmart
Submit a tool

redaction

Strip secrets, tokens, raw person properties, private URLs, and internal admin links before saving any triage output to disk or copying into customer-facing…

Skill by c-mongan

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/c-mongan-redaction

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Strip secrets, tokens, raw person properties, private URLs, and internal admin links before saving any triage output to disk or copying into customer-facing text. Use this skill before Phase 3 of the triage workflow, before posting to any channel, and before sharing any saved report outside the immediate operator.

Use cases: documents, security, productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/redaction

What's inside

Skill name
redaction
Description
Strip secrets, tokens, raw person properties, private URLs, and internal admin links before saving any triage output to disk or copying into customer-facing text. Use this skill before Phase 3 of the triage workflow, before posting to any channel, and before sharing any saved report outside the immediate operator.
License
MIT
Instructions
74 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Redaction

Saved triage reports may sit on disk for months. Customer-facing drafts get pasted into Zendesk, GitHub, email, and Slack. Both are easy places to leak secrets, PII, and internal-only links. This skill is the last line of defense.

## Categories

### Always redact

| Category | Examples | Replacement |
|---|---|---|
| API keys / tokens | `Bearer ...`, `phx_...`, `sk_...`, `ph_...`, JWT, AWS keys | `[redacted-token]` |
| Personal API keys | Any `personal_api_key`, `pat-...` | `[redacted-token]` |
| Passwords / secrets | Any field labeled `password`, `secret`, `client_secret` | `[redacted-secret]` |
| Raw person properties | Email, IP, full name, phone, address | `[redacted-pii]` |
| Session replay URLs containing IDs | `/replay/<session-id>` | `/replay/[redacted]` |
| Admin / go links | `go/admin...`, internal `/admin/...` URLs | `[admin link redacted]` |
| Private webhook URLs | Customer-side `https://<random>.ngrok.io/...` | `[redacted-private-url]` |
| Stack traces with user data | Any frame containing PII or tokens | redact the line, keep the file:line |

### Keep

These are safe and often necessary for investigation continuity:

- Public bug-tracker URLs (`https://github.com/OWNER/REPO/issues/N`).
- Public docs URLs.
- Distinct IDs and project IDs (when not customer PII).
- Org names from public issues (already public).
- Error fingerprints / hashes.
- SDK names and version numbers.
- Public dashboard URLs from your own product.

### Customer-facing extra strip

Beyond the always-redact list, the Draft Customer Response must NOT contain:

- Internal tool names (your bug tracker, your data warehouse).
- Raw queries (SQL, GraphQL, or any product-specific query language).
- Internal admin URLs even if `[redacted]`-suffixed — drop the link entirely.
- Reporter names from private tickets — use "the customer" or the org name.
- Any "I checked X internally" framing that reveals tooling.

## Pattern list (regex, copy-pasteable)

```text
(?i)bearer\s+[A-Za-z0-9._\-]{20,}
(?i)\b(sk|pk|ph|phx|phc|rk)_(live|test)?_?[A-Za-z0-9]{16,}
(?i)\bpat-[A-Za-z0-9_\-]{20,}
\beyJ[A-Za-z0-9_\-]{10,}\.[A-Za-z0-9_\-]{10,}\.[A-Za-z0-9_\-]{10,}\b   # JWT
\bAKIA[0-9A-Z]{16}\b                                                    # AWS access key
[a-z0-9._%+\-]+@[a-z0-9.\-]+\.[a-z]{2,}                                 # email (case-insensitive)
\b(?:\d{1,3}\.){3}\d{1,3}\b                                             # IPv4
go/admin[A-Za-z0-9_/\-]*                                                # internal admin links
```

These are starting points, not exhaustive. Add patterns for your own product's key shapes.

## Process

1. Generate the report and customer response in full first.
2. Run the patterns above against both files.
3. For each match: replace with the appropriate placeholder, or remove the line if the surrounding sentence is no longer informative.
4. Re-read the customer-facing response with one question in mind: *if this leaked to a public forum, would anything embarrass us or the customer?* If yes, redact further.
5. Write a one-line confirmation in the Evidence Pack: `Redaction pass: <N> replacements made across <files>.`

## When to break the rules

Almost never. Two narrow exceptions:

- Internal admin URLs MAY appear in the **internal-only** Evidence Pack inside the saved triage report, because the support engineer needs to click through. They must still be stripped from any customer-facing text and any chat post.
- A user identifier may stay in the report if both (a) the customer provided it themselves and (b) it appears in the customer response only as "the user with ID X you mentioned". Prefer "the affected user" wording when possible.

If you are unsure, redact. The cost of an over-redacted report is a follow-up question. The cost of a leaked secret is a security incident.

From c-mongan/support-triage-agent-template, MIT.

More from c-mongan/support-triage-agent-template

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
May 5, 2026 (5 months ago)
Last release
Oct 7, 2026 (21 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from c-mongan

Cite this page

redaction skill. Agent Mart. https://agentmart.dev/skills/c-mongan-redaction (last updated 2026-10-24)