heddle
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HEDDLE_LOOMWEAVE_COMMAND | No | The loomweave command used for SEI resolution and edge capture. Default is 'loomweave'. | loomweave |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| heddle_change_listB | List changed entities for an ingested repo and get the ready-to-call reverify/impact next actions. Local-only; may initialize .weft/heddle state. |
| changedB | List changed entities for an ingested repo and get the ready-to-call reverify/impact next actions. Local-only; may initialize .weft/heddle state. |
| heddle_entity_timeline_getB | Ordered change history for one entity ref. heddle reports only whether it resolved a SEI (sei_resolution); it never claims lineage. |
| timelineC | Ordered change history for one entity ref. heddle reports only whether it resolved a SEI (sei_resolution); it never claims lineage. |
| heddle_entity_churn_count_getC | Per-entity change-event count over an optional window (SEIs preferred). A never-observed entity returns churn_count 0, not an error. |
| churnB | Per-entity change-event count over an optional window (SEIs preferred). A never-observed entity returns churn_count 0, not an error. |
| heddle_impact_radius_getC | Downstream affected entities from stored dated snapshots, with mandatory completeness+staleness. NO_SNAPSHOT is an honest changed-set-only answer. |
| blast_radiusC | Downstream affected entities from stored dated snapshots, with mandatory completeness+staleness. NO_SNAPSHOT is an honest changed-set-only answer. |
| heddle_reverify_worklist_getC | Render the agent worklist to recheck before claiming completion. Sibling enrichment is advisory and never gates; absence is explicit, never clean. |
| reverifyC | Render the agent worklist to recheck before claiming completion. Sibling enrichment is advisory and never gates; absence is explicit, never clean. |
| heddle_edge_snapshot_captureC | Capture dated loomweave edges into heddle's local store. Mutates ONLY .weft/heddle state; never a sibling repo. loomweave command is server config. |
| capture_snapshotC | Capture dated loomweave edges into heddle's local store. Mutates ONLY .weft/heddle state; never a sibling repo. loomweave command is server config. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 12 tools
The tool set contains exact duplicates (e.g., changed and heddle_change_list, churn and heddle_entity_churn_count_get, etc.) with identical descriptions, making it impossible for an agent to distinguish between them. This severe overlap undermines any ability to reliably select the correct tool.
Naming is chaotic: half the tools use short names (blast_radius, changed, churn, reverify, timeline) while the other half use a 'heddle_' prefix with different patterns (heddle_change_list, heddle_edge_snapshot_capture, etc.). Verbs are mixed (capture, changed, get) and no consistent verb_noun pattern is followed.
The raw count of 12 tools is within a reasonable range, but half are duplicates, effectively reducing the unique tool count to 6. While not extreme, the duplication suggests poor scoping and undermines efficiency.
The set covers core operations like capturing snapshots, listing changes, getting churn/timeline/impact radius, but lacks update/delete or other lifecycle operations. The duplicates do not add new functionality, so coverage is adequate but not comprehensive.