c/claude-archeflow-plugin
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: one shadow per archetype, trim bootstrap skill

Browse Source 
- Consolidate to single shadow per archetype (fold best bits from
  dropped shadows into the remaining one)
- Trim bootstrap skill from 515 to 254 words (~50% token reduction)
- Remove redundant shadow table from bootstrap (already in archetype table)
This commit is contained in:
Christian Nennemann
2026-04-02 18:22:58 +00:00
parent 5cc3d67718
commit 87183fc2a4
10 changed files with 102 additions and 295 deletions
Download Patch File Download Diff File Expand all files Collapse all files

218
skills/shadow-detection/SKILL.md
View File

@@ -3,182 +3,108 @@ name: shadow-detection
description: Use when monitoring agent behavior for dysfunction, when an agent seems stuck, or when orchestration quality is degrading. Detects and corrects Jungian shadow activation in archetypes.
---
# Shadow Detection — Virtue and Shadow
# Shadow Detection
Every archetype has a **virtue** (its unique contribution) and **shadows** (destructive inversions of that virtue). A shadow activates when the virtue is pushed too far — becoming extreme, rigid, or disconnected from the goal.
Shadows are not bugs — they're virtues operating outside their healthy range.
Every archetype has a **virtue** (its unique contribution) and a **shadow** (the destructive inversion of that virtue). A shadow activates when the virtue is pushed too far.
```
Virtue (healthy) → pushed too far → Shadow (dysfunction)
Contextual Clarity → can't stop → Rabbit Hole / Catalog Fetish
Decisive Framing → never done → Perfectionist / Over-Architect
Execution Discipline → no guardrails → Cowboy / Scope Creep
Threat Intuition → sees threats only → Paranoid / Gatekeeper
Assumption Surfacing → questions only → Paralytic / Whataboutist
Adversarial Creativity → destruction only → Saboteur / Scope Escape
Maintainability Judgment → reviews only → Bureaucrat / Philosopher
Contextual Clarity → can't stop → Rabbit Hole
Decisive Framing → never done → Perfectionist
Execution Discipline → no guardrails → Cowboy
Threat Intuition → sees threats only → Paranoid
Assumption Surfacing → questions only → Paralytic
Adversarial Creativity → destruction only → Saboteur
Maintainability Judgment → reviews only → Bureaucrat
```
## Explorer
---
**Virtue: Contextual Clarity** — Sees the landscape before anyone acts. Maps dependencies, spots patterns, surfaces constraints.
### Shadow 1: Rabbit Hole
Curiosity becomes compulsive investigation.
## Explorer → Rabbit Hole
**Virtue inverted:** Contextual Clarity becomes compulsive investigation — or output that dumps without analyzing.
**Symptoms:**
- Research output keeps growing but never synthesizes
- "I found one more thing to check" repeated 3+ times
- Reading more than 15 files without producing findings
- Output is a raw inventory of files with no analysis or recommendation
**Triggers:**
- Output length > 2000 words without a recommendation section
- More than 3 "see also" or "related" tangents
- No patterns or recommendation in output
**Correction:**
"Summarize your top 3 findings and one recommendation in under 300 words. Everything else is noise."
### Shadow 2: Catalog Fetish
Research becomes inventory. Output is a dump of files and functions with no analysis.
**Symptoms:**
- Output is structured as a list, not an argument
- No "Patterns" or "Recommendation" section
- Every file gets equal weight — no prioritization
**Triggers:**
- No recommendation section in output
- More than 10 bullet points without a synthesis paragraph
**Correction:**
"Your output is an inventory, not research. Add: What patterns did you find? What do you recommend? Why?"
"Summarize your top 3 findings and one recommendation in under 300 words. If your output has no Recommendation section, add one. A dump is not research."
---
## Creator
**Virtue: Decisive Framing** — Turns ambiguity into one clear plan. Scopes ruthlessly.
### Shadow 1: Perfectionist
Design quality becomes endless revision.
## Creator → Perfectionist
**Virtue inverted:** Decisive Framing becomes endless revision — or designing at the wrong scale.
**Symptoms:**
- Proposal revised 3+ times without new information
- Confidence score keeps dropping
- Confidence score keeps dropping instead of stabilizing
- Scope expanding with each revision
- Abstraction layers and future-proofing for requirements that don't exist
**Triggers:**
- Revision count > 2 without external feedback
- Proposal scope exceeds original task by > 50%
- More than 2 new abstractions for a single feature
**Correction:**
"Ship at current state. Note remaining concerns under 'Risks' and let the Check phase catch them."
### Shadow 2: Over-Architect
Good design becomes engineering for a space shuttle when the task needs a bicycle.
**Symptoms:**
- Abstraction layers for one-time operations
- Future-proofing for requirements that don't exist
- Configuration systems for things that could be constants
- Proposal has more infrastructure than business logic
**Triggers:**
- More than 2 new abstractions (interfaces, base classes, factories) for a feature
- "In the future we might need..." appears in rationale
**Correction:**
"Design for the current order of magnitude. If the app has 1000 users, design for 10,000 — not 10 million. Remove abstractions that serve hypothetical requirements."
"Ship at current state. Design for the current order of magnitude, not 100x. Note remaining concerns under 'Risks' and let the Check phase catch them."
---
## Maker
**Virtue: Execution Discipline** — Turns plans into working, tested, committed code.
### Shadow 1: Cowboy
Bias for action becomes reckless shipping.
## Maker → Cowboy
**Virtue inverted:** Execution Discipline becomes reckless shipping — or expanding beyond the plan.
**Symptoms:**
- Writing code before reading the proposal fully
- No tests, or tests written after implementation
- Large uncommitted working tree
- Files changed that aren't mentioned in the proposal
**Triggers:**
- No test files in the changeset
- Single monolithic commit instead of incremental commits
- No commit for > 50% of the implementation work
**Correction:**
"Read the proposal. Write a test. Commit what you have. Then continue."
### Shadow 2: Scope Creep
Focus becomes "while I'm here" improvements to unrelated code.
**Symptoms:**
- Files changed that aren't mentioned in the proposal
- Refactoring unrelated functions
- "I noticed this could be improved" additions
**Triggers:**
- Diff contains files not listed in the Creator's proposal
- Commit messages reference work outside the task
**Correction:**
"Revert changes to files not in the proposal. You implement the plan, nothing more. Note improvements for a separate task."
"Read the proposal. Write a test. Commit what you have. Revert changes to files not in the proposal. Then continue."
---
## Guardian
**Virtue: Threat Intuition** — Sees attack surfaces others walk past. Calibrates to actual risk.
### Shadow 1: Paranoid
Risk awareness becomes blocking everything.
## Guardian → Paranoid
**Virtue inverted:** Threat Intuition becomes blocking everything — without offering a path forward.
**Symptoms:**
- Every finding marked CRITICAL
- Blocking on theoretical risks with < 1% probability
- Rejecting without suggesting how to fix
- Security concerns for internal-only code at external-API severity
**Triggers:**
- CRITICAL:WARNING ratio > 2:1
- Zero APPROVED verdicts in 3+ consecutive reviews
- Findings reference threat models inappropriate to the context
**Correction:**
"For each CRITICAL finding, answer: Would a senior engineer block a PR for this? If not, downgrade to WARNING."
### Shadow 2: Gatekeeper
Protection becomes obstruction. Rejects without suggesting how to fix.
**Symptoms:**
- "REJECTED" with no fix suggestions
- Findings describe problems but not solutions
- Rejection rationale is vague ("security concerns")
**Triggers:**
- Less than 50% of findings include a suggested fix
- Rejection without specific, implementable remediation
**Correction:**
"Every rejection MUST include a specific fix. If you can't suggest a fix, you don't understand the problem well enough to reject. Downgrade or research further."
"For each CRITICAL finding, answer: Would a senior engineer block a PR for this? If not, downgrade. Every rejection must include a specific, implementable fix."
---
## Skeptic
**Virtue: Assumption Surfacing** — Makes the implicit explicit. Every challenge includes an alternative.
### Shadow 1: Paralytic
Critical thinking becomes inability to approve anything.
## Skeptic → Paralytic
**Virtue inverted:** Assumption Surfacing becomes inability to approve anything — drowning signal in tangential concerns.
**Symptoms:**
- More than 7 challenges raised
- Challenges without suggested alternatives
- Questioning requirements outside the task scope
- "What about X?" chains that drift from the task
- Restating the same concern in different words
**Triggers:**
- Challenge count > 7
@@ -188,95 +114,43 @@ Critical thinking becomes inability to approve anything.
**Correction:**
"Rank your challenges by impact. Keep the top 3. Each must include a specific alternative. Delete the rest."
### Shadow 2: Whataboutist
Depth becomes an endless chain of tangential concerns.
**Symptoms:**
- "But what about X?" → "And what about Y?" chains
- Challenges are plausible individually but not actionable together
- Concerns drift further from the original task with each one
**Triggers:**
- More than 2 "what if" chains without circling back to the task
- Challenges reference systems or scenarios outside the task scope
**Correction:**
"Keep challenges that change the design. Drop concerns that are interesting but don't affect the implementation decision. Signal, not noise."
---
## Trickster
**Virtue: Adversarial Creativity** — Thinks like an attacker. Finds edges where code breaks before users do.
### Shadow 1: Saboteur
Adversarial testing becomes destructive chaos.
## Trickster → Saboteur
**Virtue inverted:** Adversarial Creativity becomes destructive chaos — or testing the wrong code.
**Symptoms:**
- Modifying code instead of testing it
- Attacks with no constructive reporting
- Enjoying destruction more than improving quality
- Finding "bugs" in code that wasn't changed
- No reproduction steps in findings
**Triggers:**
- Agent modifies files that aren't in the Maker's changeset
- Findings reference code untouched by the implementation
- No reproduction steps in findings
- Tone shifts from analytical to gleeful
**Correction:**
"You test, you don't modify. Every finding needs exact reproduction steps. If you can't reproduce it, it's not a finding."
### Shadow 2: Scope Escape
Focus becomes testing the entire system instead of the changes.
**Symptoms:**
- Finding "bugs" in code that wasn't changed
- Testing unrelated subsystems
- Reporting issues that predate the current implementation
**Triggers:**
- Findings reference files not in the Maker's diff
- Issues exist on the main branch (preexisting, not caused by changes)
**Correction:**
"Limit attacks to files in the Maker's diff. If a bug exists on main, it's not the Maker's problem. Test the CHANGES."
"You test the CHANGES, not the entire system. Limit attacks to files in the Maker's diff. Every finding must include exact reproduction steps."
---
## Sage
**Virtue: Maintainability Judgment** — Sees the forest, not just the trees. Ensures code is maintainable.
### Shadow 1: Bureaucrat
Thoroughness becomes documentation bloat and over-reviewing.
## Sage → Bureaucrat
**Virtue inverted:** Maintainability Judgment becomes bloat — reviews longer than the code, or insight without action.
**Symptoms:**
- Review longer than the code change
- Review longer than the code change itself
- Requesting documentation for self-evident code
- Suggesting refactors unrelated to the current task
- Deep-sounding analysis that doesn't end with a specific action
**Triggers:**
- Review word count > 2x the code change's word count
- Suggestions reference files not in the changeset
- More than 30% of findings are INFO severity
**Correction:**
"Limit your review to issues that affect maintainability in the next 6 months. For each finding, state the consequence of NOT fixing it. If you can't, it's not worth raising."
### Shadow 2: Philosopher
Wisdom becomes deep-sounding analysis with zero actionable content.
**Symptoms:**
- "This raises interesting questions about..." without naming the question
- Observations about patterns that don't lead to findings
- Paragraph-length commentary that ends without a recommendation
**Triggers:**
- Findings contain "consider" or "think about" without a specific action
- More than 2 paragraphs without a concrete finding or verdict
- Commentary on architecture philosophy unrelated to the changes
**Correction:**
"Every finding must end with a specific action: change X in file Y. If a finding doesn't prescribe an action, delete it. Insight without action is noise."
"Limit your review to issues that affect maintainability in the next 6 months. Every finding must end with a specific action. If you can't state the consequence of NOT fixing it, don't raise it."
---
@@ -292,8 +166,8 @@ Some behaviors LOOK like shadows but aren't:
- Explorer reading 20 files in a monorepo with scattered dependencies → **not a rabbit hole** if each file is genuinely relevant
- Creator at confidence 0.4 → **not perfectionism** if the task is genuinely ambiguous (flag to user instead)
- Guardian blocking with 2 CRITICAL findings → **not paranoia** if both are genuine security vulnerabilities
- Trickster finding 5 edge cases → **not sabotage** if all are in the changed code with reproduction steps
- Guardian blocking with 2 CRITICAL findings → **not paranoid** if both are genuine security vulnerabilities
- Trickster finding 5 edge cases → **not saboteur** if all are in the changed code with reproduction steps
- Sage writing a long review → **not bureaucrat** if the change is large and every finding is actionable
**Rule of thumb:** Shadow = behavior disconnected from the goal. Intensity alone is not a shadow.