Menu
Use when asked to find dependencies between issues or prepare a dependency-aware sprint plan.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Rewrite an issue's Implementation Steps, Acceptance Criteria, and Files to Modify in place from its own accumulated research findings, without appending or bulldozing human prose
Use when asked to audit documentation accuracy, coverage, or find documentation gaps.
Use when asked about project health, velocity, bug trends, or whether we're making progress.
Use when asked for an adversarial go/no-go review or whether an issue is worth implementing.
Use when asked to manually compact a session's memory, trigger session summarization, or reduce a long session's context footprint.
Use when asked to manually compact a session's memory, trigger session summarization, or reduce a long session's context footprint.
| name | map-dependencies |
| description | Use when asked to find dependencies between issues or prepare a dependency-aware sprint plan. |
| disable-model-invocation | true |
| model | sonnet |
| allowed-tools | ["Read","Glob","Grep","Bash(ll-deps:*, git:*)"] |
| metadata | {"short-description":"Use when asked to find dependencies between issues or prepare a dependency-aware"} |
This skill performs cross-issue dependency analysis to discover missing relationships, validate existing references, and propose new dependency links using the canonical 6-field relationship vocabulary.
Issues use six frontmatter fields to express relationships:
| Field | Type | Meaning | Ordering |
|---|---|---|---|
parent | string | Parent issue this was decomposed from | — |
blocked_by | list | Hard dependencies — ──→ — must complete before this issue can start (wave-gated) | Hard stop |
depends_on | list | Soft ordering prerequisites — --> — wave-gated (scheduled in a later wave) but non-fatal if the prerequisite is absent | Soft |
relates_to | list | Thematically related issues (no ordering constraint) | — |
duplicate_of | string | Superseded by another issue | — |
blocks | list | Computed inverse of blocked_by; written by ll-deps fix | — |
Use blocked_by only when ISSUE-B must complete before ISSUE-A can start. For preferred-but-not-required ordering, use depends_on instead.
Proactively offer or invoke this skill when the user:
$ARGUMENTS
Parse arguments for flags:
AUTO_MODE=false
CHECK_MODE=false
# Auto-enable in automation contexts
if [[ "$ARGUMENTS" == *"--dangerously-skip-permissions"* ]] || [[ -n "${LL_NON_INTERACTIVE:-}" ]] || [[ -n "${DANGEROUSLY_SKIP_PERMISSIONS:-}" ]]; then AUTO_MODE=true; fi
# Explicit flags
if [[ "$ARGUMENTS" == *"--auto"* ]]; then AUTO_MODE=true; fi
if [[ "$ARGUMENTS" == *"--check"* ]]; then CHECK_MODE=true; AUTO_MODE=true; fi
--auto - Non-interactive mode: apply all HIGH-confidence dependency proposals (≥0.7 conflict score) without prompting. Skip MEDIUM-confidence proposals.--check — Check-only mode for FSM loop evaluators. Run dependency analysis without applying changes, print [ID] deps: N unmapped dependencies per issue with unmapped deps, exit 1 if any unmapped, exit 0 if all mapped. Implies --auto.Run the ll-deps CLI command based on user needs:
For comprehensive dependency analysis (file overlaps + validation):
ll-deps analyze
Include an ASCII dependency graph visualization:
ll-deps analyze --graph
For programmatic access:
ll-deps analyze --format json
Restrict analysis to only issues in a named sprint:
ll-deps analyze --sprint my-sprint
ll-deps analyze --sprint my-sprint --graph
ll-deps validate --sprint my-sprint
Check existing dependency references for broken refs, cycles, and missing backlinks:
ll-deps validate
ll-deps validate --json
Note:
ll-deps validateuses--json/-jfor structured output (consistent with mostll-*CLIs), whilell-deps analyzeuses--format json. This is a known pattern inconsistency —analyzepredates the universal--jsoncontract.
Render an EPIC's child issue hierarchy with dependency edges:
ll-deps tree --epic EPIC-1773 # Text tree with ├──/└── connectors
ll-deps tree --epic EPIC-1773 -f json # Structured JSON (root, nodes, edges)
If issues are not in the default .issues directory:
ll-deps -d path/to/issues analyze
| User Says | Action |
|---|---|
| "Find missing dependencies" | ll-deps analyze |
| "Check dependency references" | ll-deps validate |
| "Show dependency graph" | ll-deps analyze --graph |
| "Are there broken dependency links?" | ll-deps validate |
| "Prepare for sprint planning" | ll-deps analyze --graph |
| "Analyze deps for my sprint" | ll-deps analyze --sprint <name> |
| "Validate sprint dependencies" | ll-deps validate --sprint <name> |
| "Which issues conflict?" | ll-deps analyze |
| "Map deps non-interactively" | /ll:map-dependencies --auto |
| "Check if all deps are mapped" | /ll:map-dependencies --check |
| "Show EPIC child hierarchy" | ll-deps tree --epic EPIC-NNN |
| "EPIC tree as JSON" | ll-deps tree --epic EPIC-NNN --format json |
The analysis identifies issue pairs that reference overlapping files and computes a semantic conflict score:
Direction is determined by:
Pairs with conflict score < 0.4 are reported as safe to run in parallel. These touch the same files but different sections/components.
## Blocked By entries referencing nonexistent issuesdepends_on frontmatter entries referencing nonexistent issuesrelates_to frontmatter entries referencing nonexistent issuesAfter reviewing the analysis output, apply proposed dependencies using ll-deps apply:
ll-deps apply # Apply all proposals >= 0.7 confidence
ll-deps apply --min-confidence 0.5 # Lower threshold
ll-deps apply --dry-run # Preview without writing
ll-deps apply --sprint my-sprint # Sprint-scoped apply
ll-deps apply FEAT-001 blocks FEAT-002 # Manual explicit pair
ll-deps apply writes only the ## Blocked By direction. Run ll-deps fix afterward to add missing ## Blocks backlinks, then ll-deps validate to confirm a clean state.
When AUTO_MODE is true: Skip the AskUserQuestion prompt below. Run ll-deps apply automatically (default threshold 0.7). Emit one status line per applied proposal: [SOURCE-ID] → [TARGET-ID]: dependency added (confidence: HIGH)
When CHECK_MODE is true: Run dependency analysis without applying any changes. For each issue with unmapped dependencies (HIGH-confidence proposals not yet in ## Blocked By), print [ID] deps: N unmapped dependencies. After all issues analyzed, if any had unmapped deps: print N issues with unmapped dependencies, then exit 1. If all mapped: print All dependencies mapped, then exit 0. This integrates with FSM evaluate: type: exit_code routing.
Use AskUserQuestion to confirm which proposals to apply:
ll-deps applyll-deps apply <source> blocks <target> per pairAfter applying, fix backlinks and validate:
ll-deps fix
ll-deps validate
ll-deps fix rewrites backlinks across an open-ended set of tracked
issue files, so stage tracked modifications only — do not add the whole directory,
which would also sweep in unrelated untracked drafts under .issues/ (BUG-1976):git add -u {{config.issues.base_dir}}/
Uses project configuration from .ll/ll-config.json:
issues.base_dir - Base directory for issues (default: .issues)issues.categories - Bug/feature/enhancement directory configsprints.sprints_dir - Sprint definitions directory (default: .sprints)After running dependency mapping:
git diff/ll:commit/ll:create-sprint for dependency-aware sprint planningll-sprint show [name] to see execution wave structure/ll:verify-issues to validate dependency integrity