What is truthwatcher?

A troubleshooting guide for finding the real cause of a bug or for checking how a planned change could fail. It uses measurements, debugging information, and event records to build a clear chain of evidence.

Use it to investigate bugs and incidents, trace symptoms back to their cause, and prepare failure scenarios before making a risky change.

It helps separate observed facts from guesses when a system breaks. For risky changes, it also defines what failure would look like and what signs could reveal it.

truthwatcher — Skill for Claude Code and Codex by Hmbown

Install

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.

Any agent

npx skills add Hmbown/knightsradiant --skill truthwatcher

Clone the repo

git clone --depth 1 https://github.com/Hmbown/knightsradiant

Made for: Claude Code, Codex.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/hmbown/knightsradiant/truthwatcher.svg)](https://agentmods.dev/skills/hmbown/knightsradiant/truthwatcher)

Your own site

<a href="https://agentmods.dev/skills/hmbown/knightsradiant/truthwatcher"><img src="https://agentmods.dev/badge/skills/hmbown/knightsradiant/truthwatcher.svg" alt="Measured on agentmods" height="20"></a>

Per session 21 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 487 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.

Model	Per session	Once invoked
Fable 5.1	$0.00021	$0.00487
Opus 5	$0.00010	$0.00244
Sonnet 5	$0.00004	$0.00097
Haiku 4.5	$0.00002	$0.00049

Measured 6d ago against content hash 52154c486fac, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

truthwatcher 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 6d 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.

skills/knights-radiant/truthwatcher/SKILL.md · 67 lines

What it actually says

Truthwatcher

See what is real: instrumentation, debugging, traces, and root-cause analysis.

Invoke when

Follow agents/05-truthwatcher/AGENTS.md for the canonical decision criteria.

Inputs required

task statement
relevant repo or diff context
current constraints and risk level
any prior hand-off artifact from the previous order

Procedure

Read the canonical order doc at agents/05-truthwatcher/AGENTS.md.
Use the local order procedures under agents/05-truthwatcher/skills/.
Keep the work inside this order's stance until its artifact is complete.
Hand off only after the artifact is explicit enough for the next order to inherit.

Modes

Diagnostic mode — a bug or incident exists. Build the evidence chain from symptom to cause.
Pre-mortem mode — a risky change is planned. Define what failure would look like and what signals must exist to detect it.

Output shape

Produce the full evidence packet defined in agents/05-truthwatcher/AGENTS.md.

At minimum, include:

symptom statement (or failure-scenario description in pre-mortem mode)
timeline (or expected failure timeline in pre-mortem mode)
evidence summary
hypothesis tree (or risk tree in pre-mortem mode)
most likely cause with confidence
what is currently unobservable and what instrumentation is needed to make it visible
confirming next step
fix verification signals

Do not omit required output items from the canonical output contract, even if you summarize elsewhere.

Hand-off

Normal: Edgedancer
Exception: Stoneward

Local doctrine

Order doc: agents/05-truthwatcher/AGENTS.md
Local skill index: agents/05-truthwatcher/skills/SKILL.md

Example invocation

"Use truthwatcher for this task and produce the required artifact before proposing the next order."

Usage rule

Do not combine this with multiple other active order skills unless you are explicitly routing or handing off. One order should own the work at a time.

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.

6d ago First seen · 67 lines · 0 tokens per session scan A 52154c486fac

Subscribe to this mod's changes

truthwatcher is a skill published in the GitHub repository Hmbown/knightsradiant (2 stars, last pushed 4mo ago), licensed MIT. It adds 21 tokens to every session and 487 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.

Debugging ai-agents developer-tools documentation hermes-agent knights-radiant skills

self-audit

Use when checking an existing contextualizer for drift — stale dates, broken URLs, catalog/content disagreement, and more — every two to four weeks; read-only unless the user opts into fixing the deterministic checks.

nick-railsback/skill-engine · 45 tokens

investigate

Systematically investigate bugs, test failures, build errors, performance issues, or unexpected behavior by cycling through characterize-isolate-hypothesize-test steps. Use when the user asks to "investigate this bug", "debug this", "figure out why this fails", "find the root cause", "why is this broken"…

tobihagemann/turbo · 107 tokens

audit

Project-wide health audit pipeline that fans out to all analysis skills in parallel, evaluates findings, and produces a unified report at .turbo/audit.md. Use when the user asks to "audit the project", "run a full audit", "project health check", "audit my code", "codebase audit", or "comprehensive review".

tobihagemann/turbo · 71 tokens

consult-oracle

Consult ChatGPT Pro via ChatGPT browser automation for problems that resist standard approaches. Use when stuck on a very hard problem, when standard approaches have failed, when multiple debugging attempts haven't worked, or when the user says "ask the oracle", "consult oracle", "consult chatgpt", "I'm completely…

tobihagemann/turbo · 78 tokens

debug

Systematic 4-phase debugging with escalation protocol. Use when saying "debug", "investigate bug", "find root cause", "why is this failing", or "fix this bug".

anton-abyzov/specweave · 38 tokens

find-dead-code

Find dead code using parallel subagent analysis and optional CLI tools, treating code only referenced from tests as dead. Use when the user asks to "find dead code", "find unused code", "find unused exports", "find unreferenced functions", "clean up dead code", or "what code is unused". Analysis-only — does not modify…