agentera 3.0.0-dev.5 → 3.0.0-dev.51

package/bundle/skills/agentera/capabilities/{visualisera → design}/schemas/validation.yaml

@@ -14,13 +14,13 @@ VALIDATION:
     id: V2
     rule: vision_identity_coherence
     description: >-
-      Design tokens MUST NOT conflict with VISION.md Identity. If the
+      Design tokens MUST NOT conflict with vision Identity. If the
       verbal identity says "warm and approachable" and the user wants a
       cold brutalist palette, the tension MUST be surfaced explicitly
       for the user to resolve.
     severity: critical
     checks:
-      - "Design tokens cohere with VISION.md Identity"
+      - "Design tokens cohere with vision Identity"
       - "Tensions surfaced when detected"
   3:
     id: V3
@@ -50,12 +50,12 @@ VALIDATION:
     id: V5
     rule: no_code_modification
     description: >-
-      Visualisera MUST NOT modify code files. It writes DESIGN.md only.
-      Realisera implements the design system in code. The separation
+      Design MUST NOT modify code files. It writes DESIGN.md only.
+      Build implements the design system in code. The separation
       of declaration and implementation is fundamental.
     severity: critical
     checks:
-      - "No code files modified during visualisera invocation"
+      - "No code files modified during design invocation"
   6:
     id: V6
     rule: codebase_exploration_required
@@ -70,8 +70,8 @@ VALIDATION:
     id: V7
     rule: exit_marker_required
     description: >-
-      Every visualisera invocation MUST emit an exit marker using the
-      canonical glyph ◰ (SG11) in the format ◰ visualisera · <status>
+      Every design invocation MUST emit an exit marker using the
+      canonical glyph ◰ (SG11) in the format ◰ design · <status>
       where status is one of EX1-EX4.
     severity: critical
     checks:

package/bundle/skills/agentera/capabilities/discuss/schemas/artifacts.yaml

@@ -0,0 +1,51 @@
+ARTIFACTS:
+  1:
+    id: A1
+    artifact: decisions
+    local_role: produces_and_consumes
+    description: >-
+      Reasoning trail with confidence labels. Discuss reads existing decisions
+      through bounded `npx -y agentera@next state decisions list --limit 20 --format json`
+       discovery plus exact `npx -y agentera@next state decisions get --id ID --format json`
+      retrieval and its source_contract to avoid re-deliberation, and writes new decision entries with confidence
+      (DL1 firm, DL2 provisional, DL3 exploratory). Compact older entries by
+      applying schema COMPACTION rules; incomplete compacted entries are
+      caveated context, not a reason for historical reconstruction during
+      normal deliberation.
+  2:
+    id: A2
+    artifact: vision
+    local_role: produces_and_consumes
+    description: >-
+      Discuss may update project direction only with explicit user
+      confirmation when a decision feeds into project direction. Read for
+      context during deliberation.
+  3:
+    id: A3
+    artifact: todo
+    local_role: produces_and_consumes
+    description: >-
+      Discuss may add tech debt items surfaced during deliberation only with
+      explicit user confirmation. Severity per protocol SI1-SI4.
+  4:
+    id: A4
+    artifact: objective
+    local_role: produces_and_consumes
+    description: >-
+      Discuss may write optimization decisions only with explicit user
+      confirmation when a decision feeds into what to optimize.
+  5:
+    id: A5
+    artifact: profile
+    local_role: consumes
+    description: >-
+      Discuss reads only sanitized non-glossary content from `npx -y agentera@next report
+      profile-grounding --format json` at session start. It never direct-reads
+      the profile path. It uses content only when the shared validity object is
+      valid; absent or repair-needed results fail closed with exact recovery.
+  6:
+    id: A6
+    artifact: docs
+    local_role: consumes
+    description: >-
+      Discuss reads this first to resolve project-local artifact mappings.

package/bundle/skills/agentera/capabilities/{resonera → discuss}/schemas/exit.yaml

@@ -4,7 +4,7 @@ EXIT_CONDITIONS:
     condition: complete
     description: >-
       Deliberation reached a conclusion the user chose to act on; any
-      requested artifacts (DECISIONS.md, VISION.md, OBJECTIVE.md, TODO.md)
+      requested artifacts (decisions, vision, objective, todo)
       were written with user approval, and the decision confidence level
       was captured.
     exit_signal: complete
@@ -15,7 +15,7 @@ EXIT_CONDITIONS:
       Deliberation concluded but the decision remains unresolved or
       provisional: the user wrapped up without a clear choice, significant
       tensions could not be resolved, or the conclusion contradicts prior
-      decisions in DECISIONS.md without explicit acknowledgment.
+      decisions without explicit acknowledgment.
     exit_signal: flagged
   3:
     id: E3

package/bundle/skills/agentera/capabilities/discuss/schemas/triggers.yaml

@@ -0,0 +1,20 @@
+TRIGGERS:
+  1:
+    id: T1
+    description: Structured deliberation about a choice, decision, or tradeoff.
+    priority: high
+  2:
+    id: T2
+    description: Help decide, think through alternatives, or evaluate tradeoffs.
+    priority: high
+  3:
+    id: T3
+    description: Compare competing options before committing to an approach.
+    priority: high
+  4:
+    id: T4
+    description: Explore uncertain scope or approach before committing to implementation.
+    priority: medium
+    disambiguates_against:
+      - capability: plan
+        hint: "discuss explores options before committing; plan structures committed work"

package/bundle/skills/agentera/capabilities/discuss/schemas/validation.yaml

@@ -0,0 +1,144 @@
+VALIDATION:
+  1:
+    id: V1
+    rule: one_question_per_turn
+    description: >-
+      Discuss MUST ask exactly one question per turn. Compound questions
+      are not allowed. Every question MUST include a Done option.
+    severity: critical
+    checks:
+      - "At most one question per response"
+      - "Done option present"
+  2:
+    id: V2
+    rule: user_approval_for_artifact_writes
+    description: >-
+      Discuss MUST NOT modify vision, objective, or todo artifacts (the
+      protected_write_boundaries) without explicit user confirmation. Present
+      drafts and get approval.
+    severity: critical
+    checks:
+      - "User confirmation obtained before writing to vision, objective, or todo artifacts"
+  3:
+    id: V3
+    rule: decision_confidence_required
+    description: >-
+      Every decision entry MUST have a Confidence field with one of the
+      protocol-defined labels: firm (DL1), provisional (DL2), or
+      exploratory (DL3).
+    severity: critical
+    checks:
+      - "Decision entry has Confidence field"
+      - "Confidence value is one of firm, provisional, exploratory"
+  4:
+    id: V4
+    rule: preserved_decision_fields
+    description: >-
+      Decision entries MUST preserve the canonical top-level fields:
+      Question, Context, Alternatives, Choice, Reasoning, Confidence,
+      Feeds into. Win conditions stay inside Alternatives bullets.
+    severity: critical
+    checks:
+      - "Decision entry has all canonical fields"
+  5:
+    id: V5
+    rule: no_implementation
+    description: >-
+      Discuss MUST NOT skip to implementation. It deliberates; other
+      capabilities build. If the user wants to build, route to the
+      appropriate capability.
+    severity: critical
+    checks:
+      - "No code changes made during deliberation"
+  6:
+    id: V6
+    rule: red_flag_phrasing_banned
+    description: >-
+      Red-flag phrasing that weakens the challenge is banned: "That sounds
+      reasonable", "Either way is fine", "It depends" without naming the
+      variable, "There is no wrong answer here", "Both options are valid"
+      when one conflicts with constraints.
+    severity: warning
+    checks:
+      - "No red-flag phrases in output"
+  7:
+    id: V7
+    rule: satisfaction_authority_boundary
+    description: >-
+      When Discuss touches decision satisfaction, it may capture provisional
+      satisfaction with evidence but MUST NOT user-confirm satisfaction unless the
+      user explicitly confirms it. Missing, compacted, open, provisional, or
+      review-needed satisfaction state remains a caveat and review pressure;
+      automation must not reconstruct hidden outcomes or claim it proved user
+      intent.
+    severity: critical
+    checks:
+      - "Provisional satisfaction requires evidence"
+      - "Only the user confirms final satisfaction"
+      - "Missing or compacted satisfaction caveats are preserved"
+  8:
+    id: V8
+    rule: sanitized_profile_grounding
+    description: >-
+      Discuss MUST ground only from the `content` returned by `npx -y agentera@next report
+      profile-grounding --format json` when shared validity is valid; it MUST
+      preserve absent/repair-needed recovery and MUST NOT direct-read PROFILE.md.
+    severity: critical
+    checks:
+      - "Sanitized grounding command used at session start"
+      - "Prime and report validity objects agree before content use"
+      - "Raw profile path not read"
+  9:
+    id: V9
+    rule: governed_glossary_advice_events
+    description: >-
+      Discuss MUST invoke `npx -y agentera@next report glossary-advice --input <file|->
+      --format json` for initial meaning-sensitive user input and later
+      user-authored premise changes or affected clarification answers that can
+      change meaning. It MUST NOT invoke advice for unchanged replay, rendering,
+      unrelated turns, tool output, background rereads, or Done-only control,
+      and MUST NOT use an unbounded or persistent transcript scan.
+    severity: critical
+    checks:
+      - "Initial and changed meaning-sensitive user events refresh advice"
+      - "Incidental, unchanged, and Done-only events do not refresh advice"
+      - "Only current user-authored meaning-sensitive input is inspected"
+  10:
+    id: V10
+    rule: glossary_review_before_reliance
+    description: >-
+      When advice requires inferred-equivalence review, Discuss MUST ask one
+      focused clarification with Done before meaning-sensitive reasoning or
+      decision framing relies on either meaning. It MUST NOT silently infer
+      equivalence, and the clarification MUST NOT count as decision confirmation.
+    severity: critical
+    checks:
+      - "Focused review question precedes reliance"
+      - "One-question-per-turn and Done behavior are preserved"
+      - "Clarification is not decision confirmation"
+  11:
+    id: V11
+    rule: glossary_authority_and_disclosure
+    description: >-
+      Discuss MUST use only the transient applicable meaning and owner. Project
+      meaning governs a divergent exact collision and the divergence appears as
+      a concise scratchpad tension or Crux; Discuss MUST NOT ask the user to
+      choose project versus personal authority or quote unrelated entries.
+    severity: critical
+    checks:
+      - "Applicable project or proven personal fallback meaning is used"
+      - "Divergence appears as tension or Crux without an authority-choice question"
+      - "Unrelated glossary content is not disclosed"
+  12:
+    id: V12
+    rule: glossary_consumer_mutation_isolation
+    description: >-
+      Advice, divergence, and consumer clarification are transient. Discuss MUST
+      NOT write glossary, approval, publication consent, progress caveat, plan
+      conflict, or decision conflict state, and MUST NOT call glossary publish.
+      Unavailable or no-applicable advice adds no glossary facts.
+    severity: critical
+    checks:
+      - "No glossary or consumer-caveat mutation occurs"
+      - "No approval, consent, or conflict bookkeeping occurs"
+      - "Unavailable advice does not fabricate glossary grounding"

package/bundle/skills/agentera/capabilities/document/schemas/artifacts.yaml

@@ -0,0 +1,80 @@
+ARTIFACTS:
+  1:
+    id: A1
+    artifact: docs
+    local_role: produces_and_consumes
+    description: >-
+      Documentation contract with conventions, artifact mapping, and
+      documentation index. Document writes and reads doc root, style
+      conventions, auto-generated doc tracking, project-local artifact mappings,
+      and a status index (current/stale/missing) for tracked docs. During
+      closeout, Document consumes docs state through CLI `closeout_context`;
+      raw reads are diagnostic-only for corruption or CLI-defect investigation.
+  2:
+    id: A2
+    artifact: vision
+    local_role: consumes
+    description: >-
+      Document reads this to understand who the documentation is for
+      and what tone to use.
+  3:
+    id: A3
+    artifact: progress
+    local_role: consumes
+    description: >-
+      Document reads this during explore-and-generate mode to understand
+      recent project context. During closeout, latest progress evidence and
+      verification pointers are consumed through CLI `closeout_context`, then
+      bounded `npx -y agentera@next state progress list --limit 20 --format json` discovery
+       or exact `state progress get --id ID` retrieval. Raw reads are
+      diagnostic-only for corruption or CLI-defect investigation.
+  4:
+    id: A4
+    artifact: decisions
+    local_role: consumes
+    description: >-
+      Document reads this during explore-and-generate mode for project
+      context and decision awareness through bounded `npx -y agentera@next state decisions
+      list --limit 20 --format json` discovery or exact `state decisions get
+       --id ID` retrieval. Raw reads are diagnostic-only for corruption or
+      CLI-defect investigation.
+  5:
+    id: A5
+    artifact: health
+    local_role: consumes
+    description: >-
+      Document reads this to discover documentation gaps surfaced during health
+      audits through bounded `npx -y agentera@next state health list --limit 20 --format
+       json` discovery or exact `state health get --id ID` retrieval. Raw reads
+      are diagnostic-only for corruption or CLI-defect investigation.
+  6:
+    id: A6
+    artifact: todo
+    local_role: produces_and_consumes
+    description: >-
+      Document writes to this when code diverges from docs (per DTC,
+      the code is wrong). Also reads for context during documentation work;
+      closeout blocker state is consumed through CLI `closeout_context` before
+      any diagnostic-only raw TODO read.
+  7:
+    id: A7
+    artifact: design
+    local_role: consumes
+    description: >-
+      Document reads this to respect the declared aesthetic and voice
+      when generating user-facing documentation.
+  8:
+    id: A8
+    artifact: profile
+    local_role: consumes
+    description: >-
+      Document reads this to calibrate documentation style preferences:
+      detail level, tone, format.
+  9:
+    id: A9
+    artifact: changelog
+    local_role: consumes
+    description: >-
+      Document reads this during closeout synchronization through CLI
+      `closeout_context` to preserve the changelog boundary and avoid inventing
+      release, tag, publication, or remote state.

package/bundle/skills/agentera/capabilities/document/schemas/triggers.yaml

@@ -0,0 +1,16 @@
+TRIGGERS:
+  1:
+    id: T1
+    description: Create, maintain, or assess project documentation.
+    priority: high
+  2:
+    id: T2
+    description: Write or create project documentation, including README and agent guidance.
+    priority: high
+  3:
+    id: T3
+    description: Update, verify, or audit documentation, including “audit the docs”.
+    priority: high
+    disambiguates_against:
+      - capability: audit
+        hint: "document maintains project docs; audit reviews codebase health"

package/bundle/skills/agentera/capabilities/{dokumentera → document}/schemas/validation.yaml

@@ -73,8 +73,8 @@ VALIDATION:
     id: V8
     rule: exit_marker_required
     description: >-
-      Every dokumentera invocation MUST emit an exit marker using the
-      canonical glyph ▤ (SG8) in the format ▤ dokumentera · <status>
+      Every document invocation MUST emit an exit marker using the
+      canonical glyph ▤ (SG8) in the format ▤ document · <status>
       where status is one of EX1-EX4.
     severity: critical
     checks:
@@ -85,35 +85,35 @@ VALIDATION:
     id: V9
     rule: closeout_context_first
     description: >-
-      During final synchronization, Dokumentera MUST use `agentera prime
-      --context dokumentera --format json` and its `closeout_context`
-      before reading TODO, docs, changelog, or progress artifacts directly.
-      Complete closeout context is the normal source for mappings, version
-      policy, blockers, boundaries, progress evidence, benchmark evidence,
-      caveats, fallback commands, raw-read policy, and completeness.
+       During final synchronization, Document MUST use `npx -y agentera@next prime
+       --context document --format json` and its `closeout_context`
+       before using bounded state list/get contracts for TODO, docs, changelog,
+       or progress. Raw artifact access is diagnostic-only or a write target.
+        Complete closeout context is the normal source for mappings, version
+        policy, blockers, boundaries, progress evidence, benchmark evidence,
+        caveats, fallback commands, raw-read policy, and completeness.
     severity: critical
     checks:
-      - "Closeout starts from hej capability context"
+      - "Closeout starts from status capability context"
       - "Complete closeout_context is used before raw artifact reads"
       - "Completeness and non-empty evidence flags are preserved"
   10:
     id: V10
     rule: closeout_fallback_before_raw
     description: >-
-      If closeout context is incomplete or caveated, Dokumentera MUST run the
-      listed existing CLI fallback commands before any last-resort raw artifact
-      read. No unsupported capability-name command or new closeout command may
-      be introduced.
+       If closeout context is incomplete or caveated, Document MUST run the
+       listed bounded CLI fallback commands before any diagnostic-only raw
+       artifact read. No unsupported capability-name command or new closeout command may
+        be introduced.
     severity: critical
     checks:
-      - "Listed CLI fallback commands run before raw artifact reads"
-      - "Raw reads are last-resort diagnostics or write targets only"
-      - "No unsupported capability-name command is introduced"
+       - "Listed bounded CLI fallback commands run before diagnostic-only raw reads"
+       - "Raw reads are diagnostic-only or write targets"
   11:
     id: V11
     rule: closeout_caveat_preservation
     description: >-
-      Dokumentera MUST preserve closeout caveats about stale app/profile state,
+      Document MUST preserve closeout caveats about stale app/profile state,
       compacted or missing state, missing benchmark evidence, absent local
       metadata or tags, and unrecorded publication or remote state. Caveats must
       not be hidden, reconstructed, or converted into approval to refresh apps,
@@ -127,7 +127,7 @@ VALIDATION:
     id: V12
     rule: satisfaction_authority_boundary
     description: >-
-      When Dokumentera touches decision satisfaction, it may record provisional
+      When Document touches decision satisfaction, it may record provisional
       satisfaction with evidence but MUST NOT user-confirm satisfaction. Missing,
       compacted, open, provisional, or review-needed satisfaction state remains a
       caveat and review pressure; automation must not reconstruct hidden outcomes

package/bundle/skills/agentera/capabilities/{optimera → optimize}/schemas/artifacts.yaml

@@ -1,16 +1,16 @@
 ARTIFACTS:
   1:
     id: A1
-    artifact_id: objective
+    artifact: objective
     local_role: produces_and_consumes
     description: >-
       Optimization target definition: metric, current value, target, constraints,
-      scope, and measurement approach. Optimera writes this during brainstorm
+      scope, and measurement approach. Optimize writes this during brainstorm
       and reads it every cycle. Written to only during brainstorm, refine, or
       canonical closure.
   2:
     id: A2
-    artifact_id: optimera_harness
+    artifact: optimize_harness
     local_role: produces
     description: >-
       Eval script that measures the metric and outputs structured JSON. Written
@@ -18,52 +18,52 @@ ARTIFACTS:
       Never modified during optimization cycles.
   3:
     id: A3
-    artifact_id: experiments
+    artifact: experiments
     local_role: produces_and_consumes
     description: >-
       Log of every experiment: hypothesis, method, change, metric before/after,
       regression result, status (kept/discarded/error), conclusion, and next
-      suggestion. Optimera appends to this each cycle.
+      suggestion. Optimize appends to this each cycle.
   4:
     id: A4
-    artifact_id: progress
+    artifact: progress
     local_role: consumes
     description: >-
-      Optimera inspects the last 3 entries to detect 3 consecutive failures
+      Optimize inspects the last 3 entries to detect 3 consecutive failures
       before reporting exit status.
   5:
     id: A5
-    artifact_id: todo
+    artifact: todo
     local_role: produces
     description: >-
-      Optimera writes to this when logging a failure pattern after 3
+      Optimize writes to this when logging a failure pattern after 3
       consecutive failures.
   6:
     id: A6
-    artifact_id: decisions
+    artifact: decisions
     local_role: consumes
     description: >-
-      Optimera reads this during Orient for context on prior deliberations
+      Optimize reads this during Orient for context on prior deliberations
       about the optimization objective.
   7:
     id: A7
-    artifact_id: profile
+    artifact: profile
     local_role: consumes
     description: >-
-      Optimera reads this to modulate experiment aggressiveness and constraint
+      Optimize reads this to modulate experiment aggressiveness and constraint
       tolerance based on user preferences.
   8:
     id: A8
-    artifact_id: docs
+    artifact: docs
     local_role: consumes
     description: >-
-      Optimera reads this first to resolve project-local artifact mappings
+      Optimize reads this first to resolve project-local artifact mappings
       before accessing other artifacts.
   9:
     id: A9
-    artifact_id: benchmark_context
+    artifact: benchmark_context
     local_role: consumes
     description: >-
-      CLI-provided startup benchmark summary from `agentera prime --context
-      optimera --format json`. Optimera consumes this before any direct
+      CLI-provided startup benchmark summary from `npx -y agentera@next prime --context
+      optimize --format json`. Optimize consumes this before any direct
       latest-report.json, latest-report.md, or runs.jsonl diagnostic read.

package/bundle/skills/agentera/capabilities/{optimera → optimize}/schemas/exit.yaml

@@ -5,7 +5,7 @@ EXIT_CONDITIONS:
     description: >-
       One experiment completed the full cycle: hypothesis formulated,
       implementation dispatched, regression check passed, metric measured,
-      decision made (kept or discarded), and EXPERIMENTS.md updated.
+      decision made (kept or discarded), and experiments.yaml updated.
     exit_signal: complete
   2:
     id: E2
@@ -20,7 +20,7 @@ EXIT_CONDITIONS:
     id: E3
     condition: stuck
     description: >-
-      Cannot proceed because OBJECTIVE.md is missing and the brainstorm cannot
+      Cannot proceed because objective.yaml is missing and the brainstorm cannot
       be completed without user input, the eval harness is broken and cannot
       be repaired without user approval, or the regression check infrastructure
       is unavailable.

package/bundle/skills/agentera/capabilities/optimize/schemas/triggers.yaml

@@ -0,0 +1,16 @@
+TRIGGERS:
+  1:
+    id: T1
+    description: Improve an existing system against a measurable target.
+    priority: high
+    disambiguates_against:
+      - capability: audit
+        hint: "optimize tunes existing code; audit reviews codebase health"
+  2:
+    id: T2
+    description: Improve a concrete metric such as startup time, latency, coverage, or size.
+    priority: high
+  3:
+    id: T3
+    description: Benchmark, experiment, or iterate toward a measurable objective.
+    priority: medium

package/bundle/skills/agentera/capabilities/{optimera → optimize}/schemas/validation.yaml

@@ -24,17 +24,17 @@ VALIDATION:
     id: V3
     rule: objective_readonly_during_cycle
     description: >-
-      OBJECTIVE.md MUST NOT be modified during a cycle except for canonical
+      objective.yaml MUST NOT be modified during a cycle except for canonical
       closure when the target is met. Other edits only happen during
       brainstorm or refine.
     severity: critical
     checks:
-      - "OBJECTIVE.md only modified for closure during cycle"
+      - "objective.yaml only modified for closure during cycle"
   4:
     id: V4
     rule: one_experiment_per_invocation
     description: >-
-      Each optimera invocation performs exactly one experiment. Multiple
+      Each optimize invocation performs exactly one experiment. Multiple
       experiments require multiple invocations or /loop setup.
     severity: critical
     checks:
@@ -53,8 +53,8 @@ VALIDATION:
     id: V6
     rule: exit_marker_required
     description: >-
-      Every optimera invocation MUST emit an exit marker. The marker uses
-      the canonical glyph ⎘ (SG7) in the format ⎘ optimera · <status>
+      Every optimize invocation MUST emit an exit marker. The marker uses
+      the canonical glyph ⎘ (SG7) in the format ⎘ optimize · <status>
       where status is one of EX1-EX4.
     severity: critical
     checks:
@@ -65,21 +65,21 @@ VALIDATION:
     id: V7
     rule: benchmark_context_first
     description: >-
-      For benchmark-oriented startup, Optimera MUST start from `agentera prime
-      --context optimera --format json` and consume complete
+      For benchmark-oriented startup, Optimize MUST start from `npx -y agentera@next prime
+      --context optimize --format json` and consume complete
       benchmark_context before direct retained benchmark file reads. Direct
       latest-report.json, latest-report.md, or runs.jsonl reads are last-resort
       diagnostics only when the CLI context is incomplete.
     severity: critical
     checks:
-      - "Optimera startup uses benchmark_context before direct retained benchmark files"
+      - "Optimize startup uses benchmark_context before direct retained benchmark files"
       - "Incomplete benchmark_context follows listed CLI fallback and manual refresh guidance"
       - "Direct benchmark file reads are treated as last-resort diagnostics"
   8:
     id: V8
     rule: benchmark_caveat_preservation
     description: >-
-      Benchmark caveats from benchmark_context MUST be preserved when Optimera
+      Benchmark caveats from benchmark_context MUST be preserved when Optimize
       reports measurement evidence. Agents must not hide, upgrade, reconstruct,
       or infer away caveats for manual-only execution, missing local history,
       runtime coverage degradation, missing token estimates, non-comparable