Agent skill

Coot Unmodelled Blobs

by pemsley in pemsley/coot

“How to handle Unmodelled Density Blobs”

— description from SKILL.md by pemsley
GPL-3.0Auto-check passed

Install Coot Unmodelled Blobs

skills CLI
$ npx skills add pemsley/coot --skill coot-unmodelled-blobs -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install pemsley/coot coot-unmodelled-blobs --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/pemsley/coot.git skills-src && mkdir -p .claude/skills && cp -r skills-src/mcp/docs/skills/blobs .claude/skills/coot-unmodelled-blobs && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
coot-unmodelled-blobs
GitHub stars
168
Token cost
~1.5k tokens
SKILL.md length
236 words
Files
1
Skills in repo
11
Repo updated
First seen
Licence
GPL-3.0

At a glance

  • Works in 8 steps: Go to the blob - set_rotation_centre() → Find closest atoms - not just closest… → Check if near chain terminus - look for… → …
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

About this skill

Coot Unmodelled Blobs is a skill in pemsley/coot (168 stars). Its SKILL.md is about 1.5k tokens. Licence: GPL-3.0.

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Go to the blob - set_rotation_centre()
  2. Find closest atoms - not just closest residue CAs
  3. Check if near chain terminus - look for C, O atoms of last residue
  4. If near terminus: extend chain with add_terminal_residue(), don't add OXT
  5. If near side chain: might be missing atoms, use fill_partial_residue()
  6. If isolated: might be water, ion, or ligand
  7. Always checkpoint first - make_backup_checkpoint()
  8. Refine after changes - check if blob score decreases

What it can do on your machine

Read from SKILL.md and the folder at commit 9648092. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Coot Unmodelled Blobs loads about 1.5k tokens when it runs. Until then it costs about 15 tokens; SKILL.md has 236 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~15
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from pemsley/coot at commit 9648092, republished under its GPL-3.0 licence (© pemsley). 236 words, ~1,512 tokens.

Download SKILL.mdSave it as .claude/skills/coot-unmodelled-blobs/SKILL.md (or your agent's skills folder).
name
coot-unmodelled-blobs
description
How to handle Unmodelled Density Blobs

Handling Unmodeled Density Blobs

Investigating Blobs

When find_blobs_py() identifies unmodeled density, investigate what's actually there before acting:

python
# Find blobs in difference map (molecule 2) at 3 sigma
blobs = coot.find_blobs_py(0, 2, 3.0)
# Returns: [[[x, y, z], score], ...]
# Higher score = larger/stronger blob

# Go to the biggest blob
if blobs:
    biggest = blobs[0]
    pos, score = biggest[0], biggest[1]
    coot.set_rotation_centre(pos[0], pos[1], pos[2])
Determining What a Blob Represents

Don't just look at which residues are "nearby" by CA distance - this can be misleading for non-spherical blobs.

Instead, find which atoms are closest to the blob centre:

python
blob_x, blob_y, blob_z = 59.92, 3.06, -4.23

import math
def dist(x1, y1, z1, x2, y2, z2):
    return math.sqrt((x2-x1)**2 + (y2-y1)**2 + (z2-z1)**2)

# Check all atoms near the blob
nearby_atoms = []
for chain in ['A', 'B']:
    for resno in range(1, 150):
        atoms = coot.residue_info_py(0, chain, resno, "")
        if atoms:
            res_name = coot.residue_name_py(0, chain, resno, "")
            for atom in atoms:
                x, y, z = atom[2]
                d = dist(blob_x, blob_y, blob_z, x, y, z)
                if d < 5.0:
                    atom_name = atom[0][0].strip()
                    nearby_atoms.append((d, chain, resno, res_name, atom_name))

nearby_atoms.sort()
for d, chain, resno, res_name, atom_name in nearby_atoms[:10]:
    print(f"{d:.1f}Å: {chain}/{resno} {res_name} {atom_name}")
Chain Extension vs Terminal Atoms

Critical distinction:

If a blob is near the C atom and O atom of the last residue in a chain, it usually means:

  • ❌ NOT a missing OXT (terminal carboxyl oxygen)
  • ✅ More residues to build - the chain continues!

During model building, incomplete chains are common. The density beyond the last modeled residue represents unbuilt residues, not terminal atoms.

Wrong approach:

python
# Don't do this for chain extension!
coot.add_OXT_to_residue(0, "A", 93, "")  # Wrong!

Correct approach:

python
# Extend the chain by adding residues
checkpoint = coot.make_backup_checkpoint(0, "before chain extension")

# Add residues one at a time, refining as you go
coot.add_terminal_residue(0, "A", 93, "auto", 1)  # Adds residue 94
coot.refine_residues_py(0, [["A", 93, ""], ["A", 94, ""]])
coot.accept_moving_atoms_py()

# Check if blob is still there
blobs = coot.find_blobs_py(0, 2, 3.0)
# If blob persists (maybe smaller), add another residue

coot.add_terminal_residue(0, "A", 94, "auto", 1)  # Adds residue 95
coot.refine_residues_py(0, [["A", 94, ""], ["A", 95, ""]])
coot.accept_moving_atoms_py()

# Continue until blob is gone or no more density
Iterative Chain Extension Workflow
python
def extend_chain_into_density(imol, chain_id, last_resno, imol_map, max_residues=10):
    """Extend a chain into unmodeled density."""
    
    checkpoint = coot.make_backup_checkpoint(imol, f"before extending {chain_id}")
    
    current_resno = last_resno
    residues_added = 0
    
    for i in range(max_residues):
        # Check for remaining blob near current terminus
        blobs = coot.find_blobs_py(imol, imol_map, 3.0)
        if not blobs:
            break
            
        # Get position of current C-terminus
        atoms = coot.residue_info_py(imol, chain_id, current_resno, "")
        c_pos = None
        for atom in atoms:
            if atom[0][0].strip() == "C":
                c_pos = atom[2]
                break
        
        if not c_pos:
            break
            
        # Check if any blob is near the C-terminus
        blob_near_terminus = False
        for blob in blobs:
            pos = blob[0]
            d = ((pos[0]-c_pos[0])**2 + (pos[1]-c_pos[1])**2 + (pos[2]-c_pos[2])**2)**0.5
            if d < 6.0:  # Within 6 Angstroms
                blob_near_terminus = True
                break
        
        if not blob_near_terminus:
            break
        
        # Add next residue
        result = coot.add_terminal_residue(imol, chain_id, current_resno, "auto", 1)
        if result != 1:
            break
            
        current_resno += 1
        residues_added += 1
        
        # Refine the new region
        specs = [[chain_id, r, ""] for r in range(current_resno - 2, current_resno + 1) 
                 if r > 0]
        coot.refine_residues_py(imol, specs)
        coot.accept_moving_atoms_py()
        
        # Navigate to see progress
        coot.set_go_to_atom_chain_residue_atom_name(chain_id, current_resno, "CA")
    
    print(f"Added {residues_added} residues to chain {chain_id}")
    return residues_added

# Usage:
extend_chain_into_density(0, "A", 93, 2)
When to Add OXT

Add OXT when you've finished building a fragment and there's no more density to extend into:

python
# After extending chain A as far as the density allows
# Check there's no more blob near the terminus
blobs = coot.find_blobs_py(0, 2, 3.0)
# If no blob near C-terminus, cap the chain:
coot.add_OXT_to_residue(0, "A", 96, "")
coot.refine_residues_py(0, [["A", 95, ""], ["A", 96, ""]])
coot.accept_moving_atoms_py()

The key distinction:

  • Blob near terminus → extend chain with add_terminal_residue()
  • No blob, chain fully built → cap with add_OXT_to_residue()
Summary: Blob Investigation Checklist
  1. Go to the blob - set_rotation_centre()
  2. Find closest atoms - not just closest residue CAs
  3. Check if near chain terminus - look for C, O atoms of last residue
  4. If near terminus: extend chain with add_terminal_residue(), don't add OXT
  5. If near side chain: might be missing atoms, use fill_partial_residue()
  6. If isolated: might be water, ion, or ligand
  7. Always checkpoint first - make_backup_checkpoint()
  8. Refine after changes - check if blob score decreases

© pemsley, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in mcp/docs/skills/blobs of pemsley/coot.

Open the folder on GitHubat commit 9648092

Compare with similar skills

Coot Unmodelled Blobs next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Coot Unmodelled Blobs compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Coot Unmodelled Blobs this skillpemsley/coot168—~1.5kAutomated safety check: PassGPL-3.0
Error Handlingaffaan-m/ECC277k1 repos~2.7kAutomated safety check: PassMIT
Error Handlingthedaviddias/Front-End-Checklist74k—~416Automated safety check: PassMIT
Error Handlingaffaan-m/ECC276k—~2.4kAutomated safety check: PassMIT
Python Error Handlingwshobson/agents40k—~1.5kAutomated safety check: PassMIT
Error Handling Patternswshobson/agents40k11 repos~1kAutomated safety check: PassMIT

Similar skills

  • Error Handling

    affaan-m/ECC

    Patterns for robust error handling across TypeScript, Python, and Go.

    277k GitHub starsUsed in 1 repo~2.7k tokens
    DevelopmentAuto-check passed
  • Error Handling

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing scripts, client components, bundles, or runtime behavior related to Implement proper error handling.

    74k GitHub stars~416 tokensUpdated 4 days ago
    DevelopmentAuto-check passed
  • Error Handling

    affaan-m/ECC

    TypeScript、Python、Goにわたる堅牢なエラー処理のパターン。型付きエラー、エラー境界、リトライ、サーキットブレーカー、ユーザー向けエラーメッセージをカバーします。

    276k GitHub stars~2.4k tokensUpdated yesterday
    Auto-check passed
  • Python Error Handling

    wshobson/agents

    Python error handling patterns including input validation, exception hierarchies, and partial failure handling.

    40k GitHub stars~1.5k tokensUpdated 6 days ago
    DevelopmentAuto-check passed
  • Master error handling patterns across languages including exceptions, Result types, error propagation, and graceful degradation to build resilient applications.

    40k GitHub starsUsed in 11 repos~1k tokens
    DevelopmentAuto-check passed
  • Studio Error Handling

    supabase/supabase

    Official

    Error display and troubleshooting pattern for Supabase Studio.

    111k GitHub stars~1.3k tokensUpdated yesterday
    DevelopmentAuto-check passed

More from pemsley/coot

All 11 skills in this repo
  • Coot Inline Graphs

    pemsley/coot

    Create interactive inline Chart.js graphs directly in the chat from live Coot data.

    168 GitHub stars~2.8k tokensUpdated 3 days ago
    Auto-check passed
  • Coot Rdkit

    pemsley/coot

    RDKit molecular manipulation and visualization within Coot's Python environment.

    168 GitHub stars~981 tokensUpdated 3 days ago
    Auto-check passed
  • Coot Refinement

    pemsley/coot

    Best practices for protein structure refinement and validation in Coot.

    168 GitHub stars~2k tokensUpdated 3 days ago
    Auto-check passed
  • Coot Essential API

    pemsley/coot

    API documentation to be loaded at startup - when starting a Coot session, immediately call getfunctiondescriptions() with the functions listed in this skill.

    168 GitHub stars~4.5k tokensUpdated 3 days ago
    Auto-check passed
  • Coot Figure Making

    pemsley/coot

    Best practices for creating publication-quality molecular graphics figures in Coot using user-defined colors, ribbons, and molecular representations

    168 GitHub stars~7.3k tokensUpdated 3 days ago
    Auto-check passed
  • Best Practices for Model-Building Tools and Refinement. An agent skill from pemsley/coot.

    168 GitHub stars~13k tokensUpdated 3 days ago
    Auto-check passed

Questions about Coot Unmodelled Blobs

How do I install Coot Unmodelled Blobs in Claude Code?

Run `npx skills add pemsley/coot --skill coot-unmodelled-blobs -a claude-code`. Or copy the skill folder (mcp/docs/skills/blobs in pemsley/coot) into .claude/skills/coot-unmodelled-blobs in your project. Claude Code loads it when a task matches its description.

How do I install Coot Unmodelled Blobs in Codex?

Run `npx skills add pemsley/coot --skill coot-unmodelled-blobs -a codex`. Or copy the skill folder (mcp/docs/skills/blobs in pemsley/coot) into .agents/skills/coot-unmodelled-blobs in your project. Codex loads it when a task matches its description.

Can I use Coot Unmodelled Blobs in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add pemsley/coot --skill coot-unmodelled-blobs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/coot-unmodelled-blobs, .gemini/skills/coot-unmodelled-blobs, .github/skills/coot-unmodelled-blobs and .opencode/skills/coot-unmodelled-blobs in your project.

What does Coot Unmodelled Blobs need to run?

SKILL.md names no scripts, command-line tools or credentials: Coot Unmodelled Blobs is instructions for the agent only.

Does Coot Unmodelled Blobs access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Coot Unmodelled Blobs safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Coot Unmodelled Blobs use?

Coot Unmodelled Blobs is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Coot Unmodelled Blobs use?

About 1.5k tokens (SKILL.md is roughly 6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Coot Unmodelled Blobs?

Skills that share tags, products or a category with Coot Unmodelled Blobs: Error Handling (affaan-m/ECC, 277k stars), Error Handling (thedaviddias/Front-End-Checklist, 74k stars), Error Handling (affaan-m/ECC, 276k stars) and Python Error Handling (wshobson/agents, 40k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Coot Unmodelled Blobs?

pemsley (a GitHub user) maintains it in pemsley/coot, which has 168 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on October 8, 2026.

Source: pemsley/coot on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.