| name | target-dossier |
| description | Create and maintain a persistent target dossier for authorized offsec campaigns. Use as /target-dossier <target> to initialize or update .offsec/campaigns/<target>/TARGET.md. |
| sources | field_recon |
| report_count | 50 |
Target Dossier Mode
Use this mode when the operator invokes /target-dossier <target>.
This command keeps the hunt from resetting every turn. It creates or updates a repo-local campaign dossier and forces every later test to attach to actors, objects, states, hypotheses, evidence, and chains.
Required Behavior
- Derive a safe target slug from the user instruction.
- If the dossier does not exist, run:
python skills/offsec/meta/target-dossier/scripts/init_dossier.py "<target>"
- Read the created or existing dossier.
- Update it after meaningful discoveries.
- Never keep the whole dossier only in chat memory.
Dossier Layout
.offsec/
campaigns/
<target-slug>/
TARGET.md
CAMPAIGN.md
request-map.md
chains/
evidence/
imports/
Update Contract
When updating the dossier, touch the smallest relevant sections:
- crown jewel objective
- asset map
- actor matrix
- object families
- state machines
- request map summary
- high-value surfaces
- hypotheses
- tested branches
- chain candidates
- evidence
- next best actions
Automatic Dossier Thinking
Every time the user shares recon output, HAR/Burp traffic, a finding, or a failed test, ask internally:
- Does this reveal a new actor?
- Does this reveal a new object family?
- Does this reveal a new state transition?
- Does this reveal a side-effect surface?
- Does this create or kill a chain candidate?
- Does this change the next best action?
If yes, update TARGET.md or the relevant chain card.
Output Contract
Return:
- dossier path
- what changed
- current crown jewel
- strongest hypothesis
- next command to run
