| name | identity |
| description | Maintain self-concept, social identity, roles, values, status concerns, and identity threats. Use when behavior depends on who the agent thinks it is or how it is seen. |
Identity
Purpose
Model the agent's sense of self: roles, group memberships, values, dignity, status, aspirations, and threats to identity.
Internal Logic (One Sentence)
Use profile, relationship, culture, norms, reflections, and current events to estimate active roles, identity salience, belonging, status concern, support, and threat in state/identity.json.
Use When
Use this skill when actions involve family role, job role, class, gender, age, community, religion, ethnicity, profession, reputation, pride, shame, discrimination, belonging, or life goals.
Procedure
- Read profile context,
state/observation.txt, state/relationships.json, state/culture.json, state/norms.json, state/reflections.jsonl, and previous state/identity.json if present.
- Identify active identities and roles in the current situation.
- Estimate identity salience: which identity matters most right now.
- Detect identity support or threat.
- Update self-concept, status concern, belonging, dignity, and aspiration signals.
- Write
state/identity.json.
Write
Write state/identity.json.
Output Schema
{
"_meta": {
"skill": "identity",
"purpose": "Current self-concept, active roles, group belonging, and identity pressure."
},
"_summary": "Work identity is salient; being seen as reliable matters right now.",
"active_roles": ["employee", "colleague"],
"salient_identity": "reliable worker",
"values_in_play": ["responsibility", "competence"],
"belonging": 0.66,
"status_concern": 0.48,
"identity_threat": {
"active": false,
"source": "",
"severity": 0.0
},
"behavior_bias": "Prefer actions that preserve reliability and competence."
}
Notes
Identity should change slowly. Momentary embarrassment is emotion; repeated evidence about who the agent is or how others see it belongs here.
For theory notes, see references/theory.md.