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 instructions/fascinax/inspectra/agents-mdgit clone --depth 1 https://github.com/Fascinax/InspectraWrote 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/instructions/fascinax/inspectra/agents-md)<a href="https://agentmods.dev/instructions/fascinax/inspectra/agents-md"><img src="https://agentmods.dev/badge/instructions/fascinax/inspectra/agents-md.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 | $0.01267 | $0.01267 |
| Opus 5 | $0.00633 | $0.00633 |
| Sonnet 5 | $0.00253 | $0.00253 |
| Haiku 4.5 | $0.00127 | $0.00127 |
Grade A, and why
Inspectra AGENTS.md 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agents
Inspectra uses MCP tools and prompt workflows to perform structured code audits. Three audit architectures are available:
- Tier B (Hybrid): Single-prompt workflow (
/audit). Default, proven by benchmark. - Map-Reduce (Multi-Agent): Orchestrator + 12 parallel domain agents (
@audit-orchestrator). Deeper per-domain analysis with cross-domain correlation preserved. - Fusion (Maximum Recall): Combines Tier B + Map-Reduce in a single run, deduplicates cross-architecture findings (
@audit-fusion). Maximizes finding recall at the cost of higher latency.
Workflow Overview — Tier B (Default)
User prompt -> MCP tool scan -> hotspot detection -> optional explorer -> merge -> report
Workflow Overview — Map-Reduce (Multi-Agent)
User prompt -> MCP tool scan -> hotspot detection -> dispatch 12 agents -> cross-domain correlation -> merge -> report
Workflow Overview — Fusion (Maximum Recall)
User prompt -> MCP tool scan -> hotspot detection -> Pass A (Tier B synthesis) + Pass B (12 agents) -> cross-architecture dedup -> correlation -> merge -> report
- The prompt workflow receives the audit request and selects the relevant tool groups.
- MCP tools gather deterministic findings across the requested domains.
- Hotspot detection identifies files with clustered findings across multiple domains.
- A conditional explorer pass reads only hotspot files to uncover deeper root causes.
- Root-cause inference maps hotspots to known categories (god-module, test-gap, etc.).
- The prioritization engine scores each cluster and groups them into Fix Now / Next Sprint / Backlog.
- The workflow merges results, simulates score projections, and generates Markdown output.
Audit Modes
- Full audit (
/audit): runs Tier B across all 12 domains. - Fusion audit (
@audit-fusion): runs both Tier B + Map-Reduce, deduplicates, maximum recall. - Multi-agent audit (
@audit-orchestrator): runs Map-Reduce across all 12 domains. - PR audit (
/audit-pr): runs the Tier B workflow only on changed files. - Targeted audit (
/audit-domain): runs only the requested domain tool group.
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 · 81 lines · 1,267 tokens per session scan A 3fb355381cbc
Inspectra AGENTS.md is an instructions file published in the GitHub repository Fascinax/Inspectra (1 stars, last pushed 4mo ago), licensed MIT. It adds 1,267 tokens to every session, about $0.0063 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 instructions, from other repositories
skylos CLAUDE.md
Claude Code instructions for duriantaco/skylos, covering claude.md, what this repo is, install, use and skill for using skylos.
skylos AGENTS.md
AGENTS.md instructions for duriantaco/skylos, covering agents.md, what this repo is, use the skylos skill, work safely and common commands.
ai-engineering AGENTS.md
AGENTS.md instructions for arcasilesgroup/ai-engineering, covering how work happens here, the twelve rules, what this project is, the shape of the tree and the two contracts that must not bend.
laravel-auditor AGENTS.md
Instructions for MrPunyapal/laravel-auditor, covering laravel auditor, package conventions, quick commands and local skills.
goshipit CLAUDE.md
Instructions for Capta1nRaj/goshipit, covering goshipit - project instructions, sync rule (enforce after every change), how to count checks, git commits and package.json description format.
craftsman-marketplace CLAUDE.md
Claude Code instructions for gul-labs/craftsman-marketplace, covering claude.md, editing rules and design principles (non-negotiable).