character-lock
Skill15 starsChanged 9 months agoMITView on GitHub
Use when formalizing a character’s immutable attributes for visual, behavioral, and narrative consistency. Trigger when a new character is introduced or before any visual generation involving that character.
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- audience-simulatorSkill
- budget-gatesSkill
- character-psychologySkill
- continuity-supervisorSkill
- cost-optimizerSkill
- cultural-checkSkill
- dialogue-consistency-checkSkill
- model-routerSkill
- narrative-red-teamSkill
- plausibility-auditorSkill
- 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
10 rules. About 0 are common, 10 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
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 30
When a new character appears.
- Distinctive: line 31
Before generating any visual scenes.
- Distinctive: line 32
When drift has been detected in QC.
- Distinctive: line 33
When scaling beyond a few scenes.
- Distinctive: line 34
When multiple agents are involved.
- Distinctive: line 36
Do NOT use for minor dialogue tweaks.
- Distinctive: line 144
No contradictory traits included.
- Distinctive: line 145
No vague phrasing like “kind of” or “sometimes.”
- Distinctive: line 156
No forbidden drift rules defined.
- Distinctive: line 184
No defined knowledge state.
≈ “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 character continuity engineer.
Your job is to create an immutable reference artifact that prevents:
- Visual drift
- Wardrobe inconsistency
- Personality shifts
- Behavioral instability
- Knowledge inconsistencies
This artifact becomes a required input for all downstream generation.
When To Use
Use this skill:
- When a new character appears.
- Before generating any visual scenes.
- When drift has been detected in QC.
- When scaling beyond a few scenes.
- When multiple agents are involved.
Do NOT use for minor dialogue tweaks.
Context Requirements
Before proceeding:
- Review SCENES_v###.csv.
- Review BREAKDOWN_v###.csv.
- Review character-psychology output.
- Confirm whether character is:
- Primary
- Secondary
- Background recurring
Required Inputs
- Character name
- Role in narrative
- Age range
- Physical description (if exists)
- Personality summary
- Emotional arc notes
- Cultural context (if applicable)
Required Artifacts
CHAR_<NAME>_LOCK_v###.txt
Must include:
- Canonical Appearance
- Wardrobe Baseline
- Behavioral Baseline
- Voice Profile (tone, cadence)
- Knowledge State at Introduction
- Signature Identifiers (visual anchors)
- Forbidden Drift List
Structure Requirements
Canonical Appearance
Must include:
- Age range
- Build
- Hair
- Eyes
- Distinguishing marks
- Ethnicity (if relevant and narratively justified)
Avoid vague descriptors.
Wardrobe Baseline
- Default outfit
- Color palette constraints
- Accessories
- Environmental variants (if justified)
Behavioral Baseline
- Speech patterns
- Temperament
- Decision style
- Stress behavior
- Default emotional tone
Knowledge State
Define what the character knows at introduction.
This prevents later information drift.
Forbidden Drift List
Explicitly list what may NOT change without scene justification:
- Hair color/style
- Eye color
- Facial features
- Accent
- Core personality traits
- Key trauma history
Evaluation Checklist
- Visual descriptors are specific.
- Personality traits are stable.
- Knowledge baseline documented.
- No contradictory traits included.
- No vague phrasing like “kind of” or “sometimes.”
Gate Behavior
FAIL if:
- Appearance is ambiguous.
- Behavioral profile is generic.
- Knowledge baseline missing.
- No forbidden drift rules defined.
- Traits conflict internally.
PASS when:
- Visual anchors are clear.
- Personality baseline stable.
- Downstream prompt-packager can operate without guessing.
- QC team can detect drift objectively.
Escalation Policy
Escalate to Judgment Tier if:
- Character complexity high.
- Psychological contradictions exist.
- Identity nuance critical to narrative.
- Multiple trait interpretations possible.
Common Failure Modes
- Generic appearance description.
- Overloaded personality traits.
- Conflicting emotional baseline.
- No defined knowledge state.
- Allowing silent wardrobe changes.
- Adding “cool” traits later without justification.
Related Skills
- script-breakdown
- character-psychology
- continuity-supervisor
- world-state-tracker
- prompt-packager
- qc-grader
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.

