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.
git clone --depth 1 https://github.com/axiomantic/spellbookWrote 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/commands/axiomantic/spellbook/audit-mirage-report)<a href="https://agentmods.dev/commands/axiomantic/spellbook/audit-mirage-report"><img src="https://agentmods.dev/badge/commands/axiomantic/spellbook/audit-mirage-report/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.
<a href="https://agentmods.dev/commands/axiomantic/spellbook/audit-mirage-report"><img src="https://agentmods.dev/badge/commands/axiomantic/spellbook/audit-mirage-report.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00017 | $0.01978 |
| Opus 5 | $0.00009 | $0.00989 |
| Sonnet 5 | $0.00003 | $0.00396 |
| Haiku 4.5 | $0.00002 | $0.00198 |
Grade A, and why
audit-mirage-report 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 10d 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 — 245 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Phase 5-6: Findings Report and Output
Invariant Principles
- Machine-parseable output is mandatory - The YAML block enables downstream tools (fixing-tests) to consume findings directly.
- Severity determines fix order - Critical blocks shipping; important must be addressed; minor queued.
- Reports must be self-contained - A reader must understand every finding without re-running the audit.
Machine-Parseable YAML Block
patterns_found below and the Patterns list under Human-Readable Summary MUST each
carry one slot per ### Pattern N defined in commands/audit-mirage-analyze.md. A
pattern with no slot is a finding silently dropped from a report that looks complete.
tests/test_skills/test_audit_mirage_pattern_sync.py fails when the two drift.
---
audit_metadata:
timestamp: "2024-01-15T10:30:00Z" # ISO 8601; generate actual time
test_files_audited: 5
test_functions_audited: 47
production_files_touched: 12
summary:
total_tests: 47
solid: 31
green_mirage: 12
partial: 4
skipped_total: 3
skipped_unjustified: 2
patterns_found:
pattern_1_existence_vs_validity: 3
pattern_2_partial_assertions: 4
pattern_3_shallow_matching: 2
pattern_4_lack_of_consumption: 1
pattern_5_mocking_reality: 0
pattern_6_swallowed_errors: 1
pattern_7_state_mutation: 1
pattern_8_incomplete_branches: 4
pattern_9_skipped_tests: 2
pattern_10_strengthened_still_partial: 1
pattern_11_self_manufactured_evidence: 2
findings:
- id: "finding-1"
priority: critical # critical | important | minor
test_file: "tests/test_auth.py"
test_function: "test_login_success"
line_number: 45
pattern: 2
pattern_name: "Partial Assertions"
effort: trivial # trivial | moderate | significant
depends_on: []
blind_spot: "Login could return malformed user object and test would pass"
production_impact: "Broken user sessions in production"
- id: "finding-2"
priority: critical
test_file: "tests/test_auth.py"
test_function: "test_logout"
line_number: 78
pattern: 7
pattern_name: "State Mutation Without Verification"
effort: moderate
depends_on: ["finding-1"]
blind_spot: "Session not actually cleared, just returns success"
production_impact: "Session persistence after logout"
remediation_plan:
phases:
- phase: 1
name: "Foundation fixes"
findings: ["finding-1"]
rationale: "Other tests depend on auth fixtures"
- phase: 2
name: "Auth suite completion"
findings: ["finding-2"]
rationale: "Depends on phase 1 fixtures"
total_effort_estimate: "2-3 hours"
recommended_approach: sequential # sequential: findings have dependencies; parallel: independent findings; mixed: partial dependencies
---
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.
- 10d ago First seen · 245 lines · 17 tokens per session scan A f909b80bb156
audit-mirage-report is a command published in the GitHub repository axiomantic/spellbook (10 stars, last pushed today), licensed MIT. It adds 17 tokens to every session and 1,978 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.
Other commands, from other repositories
atomic-help
Route a lost user to the right atomic verb, skill, agent, or binary subcommand. Bare invocation reads git state and recommends one next step. Topic keyword or freeform intent → focused pointer. tour → guided walkthrough of the system. Help router, not duplicated docs.
report-issue
Open a atomic GitHub issue via gh. Bug report or feature request — auto-detect from user's description.
test-integration.en
Run the integration test workflow.
test.en
Run the full project test workflow.
pr-ready
Run the project's pre-commit review loop to determine whether the current branch is ready to push — lint, tests, parallel pr-review-toolkit agents plus an over-engineering audit, fix-and-re-run until convergence.
refactor
Refactoring command. Detects code smells and creates a safe refactoring plan.