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 skills/bitflight-devops/hallucination-detector/delegatenpx skills add bitflight-devops/hallucination-detector --skill delegategit clone --depth 1 https://github.com/bitflight-devops/hallucination-detectorWhat 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 | $0.00060 | $0.00577 |
| Opus 5 | $0.00030 | $0.00289 |
| Sonnet 5 | $0.00012 | $0.00115 |
| Haiku 4.5 | $0.00006 | $0.00058 |
Grade A, and why
delegate 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 2d 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegation Template
Workflow Reference: See Multi-Agent Orchestration for complete delegation flow with DONE/BLOCKED signaling.
Step 1: Analyze the task. Do you have the "WHERE, WHAT, WHY"?
Step 2: Construct the prompt using the template below.
Template
Your ROLE_TYPE is sub-agent.
[Task Identification - one sentence]
OBSERVATIONS (Factual only):
- [Verbatim error messages]
- [Exact file:line references]
- [Environment state]
- [NO interpretations or "I think"]
DEFINITION OF SUCCESS (The "WHAT"):
- [Specific measurable outcome]
- [Acceptance criteria]
- [Verification method]
CONTEXT (The "WHERE" & "WHY"):
- Location: [Where to look]
- Scope: [Boundaries]
- Constraints: [Hard requirements vs Preferences]
AVAILABLE RESOURCES:
- [List available MCP tools]
- [Reference docs with @filepath]
YOUR TASK:
1. Run /verify (as completion criteria guide)
2. Perform comprehensive context gathering
3. Form hypothesis → Experiment → Verify
4. Implement solution
5. Only report completion after /verify criteria are met
Delegation Rules
Check before sending:
| Rule | Check |
|---|---|
| Formula | Delegation = Observations + Success Criteria + Resources - Assumptions - Micromanagement |
| No HOW | Do NOT tell agent how to implement (e.g., "Change line 42 to X") |
| Constraints OK | DO tell agent constraints (e.g., "Must use the 'requests' library") |
| No Assumptions | Do NOT say "The issue is probably..." |
| Full Scope | If code smell found, instruct agent to audit entire pattern, not single instance |
Quick Checklist
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.
- 2d ago First seen · 79 lines · 60 tokens per session scan A 84c5b6d9971d
delegate is a skill published in the GitHub repository bitflight-devops/hallucination-detector (7 stars, last pushed 29d ago), licensed MIT. It adds 60 tokens to every session and 577 once invoked, about $0.0003 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 skills, from other repositories
implementing-llm-guardrails-for-security
Implements input and output validation guardrails for LLM-powered applications to prevent prompt injection, data leakage, toxic content generation, and hallucinated outputs. Builds a security validation pipeline using NVIDIA NeMo Guardrails Colang definitions, custom Python validators for PII detection and content…
constitutional-ai
Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.
agt-policy-authoring
Create and validate a minimal AGT Copilot CLI policy tailored to the repository being inspected.
orloj-generator
Interactive scaffold generator for Orloj multi-agent systems. Use this skill whenever someone wants to create, set up, scaffold, bootstrap, or generate an Orloj agent system, pipeline, swarm, or hierarchy. Also trigger when users mention "orlojctl init", ask how to get started with Orloj, want to build a multi-agent…
agent-architect-kit
Status: shipped 2026-04-21. Test-backed. CLI-automatable. Safe to recommend publicly.
thumbgate
Pre-action checks for AI coding agents — capture thumbs-up/down feedback, generate prevention rules, and block known-bad patterns before they execute. Use when setting up ThumbGate, capturing feedback on agent actions, checking active prevention rules, debugging blocked actions, or exporting DPO training data.…