How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

ai-collab-map - Prompt - OpenSmartRoute

Name: ai-collab-map
Author: Kedarnadhsharma

Prompt file imported from Kedarnadhsharma/engineering-management-ai-toolkit (.claude/commands/people/ai-collab-map.md). Copyright stays with the author.

/ai-collab-map — Team Collaboration & Silo Detector

Segment: Team Culture → Collaboration Trigger: Monthly (first Monday of the month)

What This Skill Does

Maps who is @-mentioning, DM-ing, and co-authoring Jira tickets with whom. Identifies silos, over-reliance patterns, and cross-functional gaps. Delivers a collaboration map with 2–3 specific actions to improve connectivity.

Silos don't announce themselves. They accumulate quietly and surface as knowledge bus factors when someone leaves.

Instructions

Ask the user for:

List of team members' Slack handles
Standup channel name and other key team channels
Partner team Slack channels (where cross-team collaboration should be visible)
Jira project key(s)
Their Slack user ID (for DM delivery)

Step 1 — Map @-mention patterns

Use mcp__slack__slack_conversation_history across standup and team channels for the past 30 days. For each message, extract @-mentions. Build a directed graph: who mentions whom.

Use mcp__slack__slack_resolve_user_id to map all handles to names.

Track:

Internal @-mentions (team member to team member)
Cross-team @-mentions (team member to someone outside the team)
Reciprocity: does @A mention @B and @B mention @A, or is it one-directional?

Step 2 — Find group DM clusters

Use mcp__slack__slack_list_mpims to identify group DM conversations involving team members. Note: who is consistently in group DMs together vs. who is excluded.

Step 3 — Map Jira co-assignment and collaboration

Use mcp__atlassian__searchJiraIssuesUsingJql:

project in (<PROJECT_KEYS>) AND updated >= -30d

For each ticket, note: assignee, reporter, watchers, and commenters. Build a co-work graph: who appears on the same tickets together.

Step 4 — Scan partner team channels

Use mcp__slack__slack_conversation_history on the partner team channels provided. Identify which of your team members are active there (cross-functional visibility) and which are absent.

Step 5 — Identify collaboration patterns

Silo signals: team members with zero @-mentions from/to others in the past 30 days, no cross-team channel activity, all Jira tickets solo-assigned.

Over-reliance signals: one person appearing as a dependency in many others' work (@-mentioned frequently by peers, many tickets where they're the sole commenter resolving blockers).

Cross-functional gaps: team members with high internal collaboration but zero presence in partner team channels or cross-team tickets.

Healthy nodes: people with balanced bidirectional interaction — giving and receiving.

Step 6 — Deliver the collaboration map

Use mcp__slack__slack_post to DM the map to the manager.

Step 7 — Save HTML output

Write a clean, self-contained HTML file named [DD-MM-YYYY].html (today's date, e.g. 06-06-2026.html) in the same folder as this skill file (.claude/commands/people/). The file should mirror the output format below with minimal styling — system font, white background, subtle card sections, and colored status indicators (green #22c55e for connected, amber #f59e0b for moderate, red #ef4444 for isolated). No external CSS or JS dependencies.

Output Format

COLLABORATION MAP — [MONTH]

INTERACTION DENSITY
• @[name]: [N] unique teammates interacted with, [N] cross-team connections
  Status: CONNECTED / MODERATE / ISOLATED
  ...

SILO SIGNALS
• @[name]: [evidence — e.g., "no @-mentions sent or received in 30 days, zero partner channel activity"]
  Risk: knowledge bus factor if they leave; limited influence growth
  → Action: [specific pairing or visibility opportunity]

OVER-RELIANCE SIGNALS
• @[name]: mentioned as a dependency by [N] teammates, [N] tickets where they're sole resolver
  Risk: single point of failure, potential burnout, blocks others when unavailable
  → Action: [distribute knowledge — pairing, documentation, shadowing]

CROSS-FUNCTIONAL GAPS
• @[name]: strong internal collaboration, zero presence in partner channels
  → Action: [specific cross-team meeting or channel to join]

HEALTHY CONNECTIONS
• @[name] ↔ @[name]: strong bidirectional collaboration
• Team cross-functional reach: [N] partner teams engaged this month

2–3 ACTIONS TO IMPROVE CONNECTIVITY
1. [Specific action — e.g., "Pair @alice with @bob on TICKET-X to build cross-team coverage"]
2. [Specific action]
3. [Specific action]

Install ai-collab-map and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# /ai-collab-map — Team Collaboration & Silo Detector **Segment:** Team Culture → Collaboration **Trigger:** Monthly (first Monday of the month) ## What This Skill Does Maps who is @-mentioning, DM-ing, and co-authoring Jira tickets with whom. Identifies silos, over-reliance patterns, and cross-functional gaps. Delivers a collaboration map with 2–3 specific actions to improve connectivity. Silos don't announce themselves. They accumulate quietly and surface as knowledge bus factors when someone leaves. ## Instructions Ask the user for: - List of team members' Slack handles - Standup channel name and other key team channels - Partner team Slack channels (where cross-team collaboration should be visible) - Jira project key(s) - Their Slack user ID (for DM delivery) ### Step 1 — Map @-mention patterns Use `mcp__slack__slack_conversation_history` across standup and team channels for the past 30 days. For each message, extract @-mentions. Build a directed graph: who mentions whom. Use `mcp__slack__slack_resolve_user_id` to map all handles to names. Track: - Internal @-mentions (team member to team member) - Cross-team @-mentions (team member to someone outside the team) - Reciprocity: does @A mention @B and @B mention @A, or is it one-directional? ### Step 2 — Find group DM clusters Use `mcp__slack__slack_list_mpims` to identify group DM conversations involving team members. Note: who is consistently in group DMs together vs. who is excluded. ### Step 3 — Map Jira co-assignment and collaboration Use `mcp__atlassian__searchJiraIssuesUsingJql`: ``` project in (<PROJECT_KEYS>) AND updated >= -30d ``` For each ticket, note: assignee, reporter, watchers, and commenters. Build a co-work graph: who appears on the same tickets together. ### Step 4 — Scan partner team channels Use `mcp__slack__slack_conversation_history` on the partner team channels provided. Identify which of your team members are active there (cross-functional visibility) and which are absent. ### Step 5 — Identify collaboration patterns **Silo signals**: team members with zero @-mentions from/to others in the past 30 days, no cross-team channel activity, all Jira tickets solo-assigned. **Over-reliance signals**: one person appearing as a dependency in many others' work (@-mentioned frequently by peers, many tickets where they're the sole commenter resolving blockers). **Cross-functional gaps**: team members with high internal collaboration but zero presence in partner team channels or cross-team tickets. **Healthy nodes**: people with balanced bidirectional interaction — giving and receiving. ### Step 6 — Deliver the collaboration map Use `mcp__slack__slack_post` to DM the map to the manager. ### Step 7 — Save HTML output Write a clean, self-contained HTML file named `[DD-MM-YYYY].html` (today's date, e.g. `06-06-2026.html`) in the same folder as this skill file (`.claude/commands/people/`). The file should mirror the output format below with minimal styling — system font, white background, subtle card sections, and colored status indicators (green `#22c55e` for connected, amber `#f59e0b` for moderate, red `#ef4444` for isolated). No external CSS or JS dependencies. ## Output Format ``` COLLABORATION MAP — [MONTH] INTERACTION DENSITY • @[name]: [N] unique teammates interacted with, [N] cross-team connections Status: CONNECTED / MODERATE / ISOLATED ... SILO SIGNALS • @[name]: [evidence — e.g., "no @-mentions sent or received in 30 days, zero partner channel activity"] Risk: knowledge bus factor if they leave; limited influence growth → Action: [specific pairing or visibility opportunity] OVER-RELIANCE SIGNALS • @[name]: mentioned as a dependency by [N] teammates, [N] tickets where they're sole resolver Risk: single point of failure, potential burnout, blocks others when unavailable → Action: [distribute knowledge — pairing, documentation, shadowing] CROSS-FUNCTIONAL GAPS • @[name]: strong internal collaboration, zero presence in partner channels → Action: [specific cross-team meeting or channel to join] HEALTHY CONNECTIONS • @[name] ↔ @[name]: strong bidirectional collaboration • Team cross-functional reach: [N] partner teams engaged this month 2–3 ACTIONS TO IMPROVE CONNECTIVITY 1. [Specific action — e.g., "Pair @alice with @bob on TICKET-X to build cross-team coverage"] 2. [Specific action] 3. [Specific action] ```

{ "prompt": "# /ai-collab-map — Team Collaboration & Silo Detector\n\n**Segment:** Team Culture → Collaboration\n**Trigger:** Monthly (first Monday of the month)\n\n## What This Skill Does\n\nMaps who is @-mentioning, DM-ing, and co-authoring Jira tickets with whom. Identifies silos, over-reliance patterns, and cross-functional gaps. Delivers a collaboration map with 2–3 specific actions to improve connectivity.\n\nSilos don't announce themselves. They accumulate quietly and surface as knowledge bus factors when someone leaves.\n\n## Instructions\n\nAsk the user for:\n- List of team members' Slack handles\n- Standup channel name and other key team channels\n- Partner team Slack channels (where cross-team collaboration should be visible)\n- Jira project key(s)\n- Their Slack user ID (for DM delivery)\n\n### Step 1 — Map @-mention patterns\nUse `mcp__slack__slack_conversation_history` across standup and team channels for the past 30 days. For each message, extract @-mentions. Build a directed graph: who mentions whom.\n\nUse `mcp__slack__slack_resolve_user_id` to map all handles to names.\n\nTrack:\n- Internal @-mentions (team member to team member)\n- Cross-team @-mentions (team member to someone outside the team)\n- Reciprocity: does @A mention @B and @B mention @A, or is it one-directional?\n\n### Step 2 — Find group DM clusters\nUse `mcp__slack__slack_list_mpims` to identify group DM conversations involving team members. Note: who is consistently in group DMs together vs. who is excluded.\n\n### Step 3 — Map Jira co-assignment and collaboration\nUse `mcp__atlassian__searchJiraIssuesUsingJql`:\n```\nproject in (<PROJECT_KEYS>) AND updated >= -30d\n```\nFor each ticket, note: assignee, reporter, watchers, and commenters. Build a co-work graph: who appears on the same tickets together.\n\n### Step 4 — Scan partner team channels\nUse `mcp__slack__slack_conversation_history` on the partner team channels provided. Identify which of your team members are active there (cross-functional visibility) and which are absent.\n\n### Step 5 — Identify collaboration patterns\n\n**Silo signals**: team members with zero @-mentions from/to others in the past 30 days, no cross-team channel activity, all Jira tickets solo-assigned.\n\n**Over-reliance signals**: one person appearing as a dependency in many others' work (@-mentioned frequently by peers, many tickets where they're the sole commenter resolving blockers).\n\n**Cross-functional gaps**: team members with high internal collaboration but zero presence in partner team channels or cross-team tickets.\n\n**Healthy nodes**: people with balanced bidirectional interaction — giving and receiving.\n\n### Step 6 — Deliver the collaboration map\nUse `mcp__slack__slack_post` to DM the map to the manager.\n\n### Step 7 — Save HTML output\nWrite a clean, self-contained HTML file named `[DD-MM-YYYY].html` (today's date, e.g. `06-06-2026.html`) in the same folder as this skill file (`.claude/commands/people/`). The file should mirror the output format below with minimal styling — system font, white background, subtle card sections, and colored status indicators (green `#22c55e` for connected, amber `#f59e0b` for moderate, red `#ef4444` for isolated). No external CSS or JS dependencies.\n\n## Output Format\n\n```\nCOLLABORATION MAP — [MONTH]\n\nINTERACTION DENSITY\n• @[name]: [N] unique teammates interacted with, [N] cross-team connections\n Status: CONNECTED / MODERATE / ISOLATED\n ...\n\nSILO SIGNALS\n• @[name]: [evidence — e.g., \"no @-mentions sent or received in 30 days, zero partner channel activity\"]\n Risk: knowledge bus factor if they leave; limited influence growth\n → Action: [specific pairing or visibility opportunity]\n\nOVER-RELIANCE SIGNALS\n• @[name]: mentioned as a dependency by [N] teammates, [N] tickets where they're sole resolver\n Risk: single point of failure, potential burnout, blocks others when unavailable\n → Action: [distribute knowledge — pairing, documentation, shadowing]\n\nCROSS-FUNCTIONAL GAPS\n• @[name]: strong internal collaboration, zero presence in partner channels\n → Action: [specific cross-team meeting or channel to join]\n\nHEALTHY CONNECTIONS\n• @[name] ↔ @[name]: strong bidirectional collaboration\n• Team cross-functional reach: [N] partner teams engaged this month\n\n2–3 ACTIONS TO IMPROVE CONNECTIVITY\n1. [Specific action — e.g., \"Pair @alice with @bob on TICKET-X to build cross-team coverage\"]\n2. [Specific action]\n3. [Specific action]\n```", "variables": [], "notes": "Maps who is @-mentioning, DM-ing, and co-authoring Jira tickets with whom. Identifies silos, over-reliance patterns, and cross-functional gaps. Delivers a collaboration map with 2–3 specific actions to improve connectivity.", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/Kedarnadhsharma/engineering-management-ai-toolkit", "path": ".claude/commands/people/ai-collab-map.md", "ref": "6e5447bcd0debfeabe748128a9c50323157ff21e", "url": "https://github.com/Kedarnadhsharma/engineering-management-ai-toolkit/blob/6e5447bcd0debfeabe748128a9c50323157ff21e/.claude/commands/people/ai-collab-map.md", "key": "Kedarnadhsharma/engineering-management-ai-toolkit/.claude/commands/people/ai-collab-map.md" } } }

ai-collab-map

About

/ai-collab-map — Team Collaboration & Silo Detector

What This Skill Does

Instructions

Step 1 — Map @-mention patterns

Step 2 — Find group DM clusters

Step 3 — Map Jira co-assignment and collaboration

Step 4 — Scan partner team channels

Step 5 — Identify collaboration patterns

Step 6 — Deliver the collaboration map

Step 7 — Save HTML output

Output Format

Install ai-collab-map and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews