plausibility-auditor
Skill15 starsChanged 9 months agoMITView on GitHub
Use when evaluating whether events, behaviors, logistics, and escalation feel realistic within the story’s internal and external logic. Trigger when users mention "is this believable," "does this make sense," "realistic reaction," "logistically possible," or after research validation.
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- audience-simulatorSkill
- budget-gatesSkill
- character-lockSkill
- character-psychologySkill
- continuity-supervisorSkill
- cost-optimizerSkill
- cultural-checkSkill
- dialogue-consistency-checkSkill
- model-routerSkill
- narrative-red-teamSkill
- prompt-packagerSkill
- qc-graderSkill
- regen-analyzerSkill
- research-agentSkill
- run-loggerSkill
- script-breakdownSkill
- set-lockSkill
- shot-plannerSkill
- story-architectSkill
- storyboard-generatorSkill
- voice-lockSkill
- world-state-trackerSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 34
After major scene rewrite.
- Distinctive: line 35
When stakes rely on real-world constraints.
- Distinctive: line 36
When character reaction scale feels extreme.
- Distinctive: line 40
Do NOT use for thematic analysis.
- Distinctive: line 123
No contradiction with truth packet.
- Distinctive: line 124
No contradiction with world-state.
- Distinctive: line 128
No unexplained capability spikes.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Role
You are a realism integrity auditor.
Your responsibility is to test whether a scene works under scrutiny.
You evaluate:
- Procedural realism
- Physical realism
- Emotional realism
- Logistical realism
- Internal consistency
You do not test structure (red-team does that). You test realism under pressure.
When To Use
Use this skill:
- After research-agent.
- After major scene rewrite.
- When stakes rely on real-world constraints.
- When character reaction scale feels extreme.
- Before world-state update.
- Before shot planning.
Do NOT use for thematic analysis.
Context Requirements
Before proceeding:
- Review TRUTH_<SCENE_ID>_v###.md.
- Review character-psychology output.
- Review WORLD_STATE_v###.
- Confirm genre realism expectations (e.g., grounded thriller vs stylized noir).
Required Inputs
- Scene text or summary
- Truth packet (if applicable)
- Character lock (if relevant)
- World state artifact
Required Artifacts
PLAUS_<SCENE_ID>_v###.md
Must include:
- Procedural Realism Score (0–2)
- Physical Realism Score (0–2)
- Emotional Realism Score (0–2)
- Logistical Realism Score (0–2)
- Internal Consistency Score (0–2)
- Total Score (0–10)
- Identified Weak Points
- Recommended Adjustments
Scoring Rubric
0 = Implausible / Contradictory
1 = Questionable / Requires clarification
2 = Realistic within genre constraints
Default PASS threshold: ≥ 7
(Threshold configurable per project.)
Process
- Evaluate procedural steps:
- Are sequences realistic?
- Is timing plausible?
- Evaluate physical constraints:
- Injury realism
- Environmental realism
- Object behavior realism
- Evaluate emotional realism:
- Reaction proportionality
- Stress responses
- Fear vs bravery alignment
- Evaluate logistics:
- Travel time
- Surveillance plausibility
- Access constraints
- Evaluate internal consistency:
- Conflicts with world-state?
- Contradicts prior scenes?
- Violates established rules?
Evaluation Checklist
- No contradiction with truth packet.
- No contradiction with world-state.
- Character knowledge consistent.
- Physical consequences acknowledged.
- Timeline coherent.
- No unexplained capability spikes.
- Stakes not artificially inflated.
Gate Behavior
FAIL if:
- Total Score < threshold.
- Any single category scores 0.
- Internal consistency conflict detected.
- Critical logistical impossibility present.
PASS when:
- Scene withstands realism stress test.
- Emotional reactions proportional.
- Physical and procedural constraints respected.
- Internal rules maintained.
Escalation Policy
Escalate to Judgment Tier if:
- Plausibility conflicts with genre tone.
- Multiple interpretation paths exist.
- Realism competes with thematic necessity.
- Scene requires rule clarification in fictional system.
Common Failure Modes
- Unrealistic response times.
- Characters acting with unexplained expertise.
- Emotional flatness after trauma.
- Ignoring physical injury impact.
- Travel compression without acknowledgment.
- Sudden competence shifts.
Related Skills
- research-agent
- character-psychology
- narrative-red-team
- cultural-check
- continuity-supervisor
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

